Datasets:

bunkalab

/

arxiv_datasets_llm

like 1

The Critique of Critique

Shichao Sun, Junlong Li, Weizhe Yuan, Ruifeng Yuan, Wenjie Li, Pengfei Liu

Critique, as a natural language description for assessing the quality of model-generated content, has been proven to play an essential role in the training, evaluation, and refinement of Large Language Models (LLMs). However, there is a lack of principled understanding in evaluating the quality of the critique itself. In this paper, we pioneer the critique of critique, termed MetaCritique, which is a framework to evaluate the critique from two aspects, i.e., factuality as precision score and comprehensiveness as recall score. We calculate the harmonic mean of precision and recall as the overall rating called F1 score. To obtain a reliable evaluation outcome, we propose Atomic Information Units (AIUs), which describe the critique in a more fine-grained manner. MetaCritique takes each AIU into account and aggregates each AIU's judgment for the overall score. Moreover, given the evaluation process involves intricate reasoning, our MetaCritique provides a natural language rationale to support each judgment. We construct a meta-evaluation dataset containing 300 critiques (2653 AIUs) across four tasks (question answering, reasoning, entailment, and summarization), and we conduct a comparative study to demonstrate the feasibility and effectiveness. Experiments also show superior critique judged by MetaCritique leads to better refinement, indicating generative artificial intelligence indeed has the potential to be significantly advanced with our MetaCritique. We will release relevant code and meta-evaluation datasets at https://github.com/GAIR-NLP/MetaCritique.

http://arxiv.org/abs/2401.04518v1

cs.CL, cs.AI

Exploring Prompt-Based Methods for Zero-Shot Hypernym Prediction with Large Language Models

Mikhail Tikhomirov, Natalia Loukachevitch

This article investigates a zero-shot approach to hypernymy prediction using large language models (LLMs). The study employs a method based on text probability calculation, applying it to various generated prompts. The experiments demonstrate a strong correlation between the effectiveness of language model prompts and classic patterns, indicating that preliminary prompt selection can be carried out using smaller models before moving to larger ones. We also explore prompts for predicting co-hyponyms and improving hypernymy predictions by augmenting prompts with additional information through automatically identified co-hyponyms. An iterative approach is developed for predicting higher-level concepts, which further improves the quality on the BLESS dataset (MAP = 0.8).

http://arxiv.org/abs/2401.04515v1

cs.CL, cs.AI

Fighting Fire with Fire: Adversarial Prompting to Generate a Misinformation Detection Dataset

Shrey Satapara, Parth Mehta, Debasis Ganguly, Sandip Modha

The recent success in language generation capabilities of large language models (LLMs), such as GPT, Bard, Llama etc., can potentially lead to concerns about their possible misuse in inducing mass agitation and communal hatred via generating fake news and spreading misinformation. Traditional means of developing a misinformation ground-truth dataset does not scale well because of the extensive manual effort required to annotate the data. In this paper, we propose an LLM-based approach of creating silver-standard ground-truth datasets for identifying misinformation. Specifically speaking, given a trusted news article, our proposed approach involves prompting LLMs to automatically generate a summarised version of the original article. The prompts in our proposed approach act as a controlling mechanism to generate specific types of factual incorrectness in the generated summaries, e.g., incorrect quantities, false attributions etc. To investigate the usefulness of this dataset, we conduct a set of experiments where we train a range of supervised models for the task of misinformation detection.

http://arxiv.org/abs/2401.04481v1

cs.CL, cs.AI

SonicVisionLM: Playing Sound with Vision Language Models

Zhifeng Xie, Shengye Yu, Qile He, Mengtian Li

There has been a growing interest in the task of generating sound for silent videos, primarily because of its practicality in streamlining video post-production. However, existing methods for video-sound generation attempt to directly create sound from visual representations, which can be challenging due to the difficulty of aligning visual representations with audio representations. In this paper, we present SonicVisionLM, a novel framework aimed at generating a wide range of sound effects by leveraging vision-language models(VLMs). Instead of generating audio directly from video, we use the capabilities of powerful VLMs. When provided with a silent video, our approach first identifies events within the video using a VLM to suggest possible sounds that match the video content. This shift in approach transforms the challenging task of aligning image and audio into more well-studied sub-problems of aligning image-to-text and text-to-audio through the popular diffusion models. To improve the quality of audio recommendations with LLMs, we have collected an extensive dataset that maps text descriptions to specific sound effects and developed a time-controlled audio adapter. Our approach surpasses current state-of-the-art methods for converting video to audio, enhancing synchronization with the visuals, and improving alignment between audio and video components. Project page: https://yusiissy.github.io/SonicVisionLM.github.io/

http://arxiv.org/abs/2401.04394v3

cs.MM, cs.SD, eess.AS

Improving the Robustness of Knowledge-Grounded Dialogue via Contrastive Learning

Jiaan Wang, Jianfeng Qu, Kexin Wang, Zhixu Li, Wen Hua, Ximing Li, An Liu

Knowledge-grounded dialogue (KGD) learns to generate an informative response based on a given dialogue context and external knowledge (\emph{e.g.}, knowledge graphs; KGs). Recently, the emergence of large language models (LLMs) and pre-training techniques has brought great success to knowledge-grounded dialogue. However, when building KGD systems in real applications, there are various real-world noises that are inevitable to face. For example, the dialogue context might involve perturbations such as misspellings and abbreviations. In addition, KGs typically suffer from incompletion and also might contain erroneous and outdated facts. Such real-world noises pose a challenge to the robustness of KGD systems and hinder their applications in the real world. In this paper, we propose an entity-based contrastive learning framework for improving the robustness of KGD. Specifically, we make use of the entity information in a KGD sample to create both its positive and negative samples which involve semantic-irrelevant and semantic-relevant perturbations, respectively. The contrastive learning framework ensures the KGD model is aware of these two types of perturbations, thus generating informative responses with the potentially noisy inputs in real applications. Experimental results on three benchmark datasets show that our method achieves new state-of-the-art performance in terms of automatic evaluation scores, verifying its effectiveness and potentiality. Furthermore, we show that our method can generate better responses than comparison models in both the noisy and the few-shot settings.

http://arxiv.org/abs/2401.04361v1

cs.CL, cs.AI

Large Language Models for Robotics: Opportunities, Challenges, and Perspectives

Jiaqi Wang, Zihao Wu, Yiwei Li, Hanqi Jiang, Peng Shu, Enze Shi, Huawen Hu, Chong Ma, Yiheng Liu, Xuhui Wang, Yincheng Yao, Xuan Liu, Huaqin Zhao, Zhengliang Liu, Haixing Dai, Lin Zhao, Bao Ge, Xiang Li, Tianming Liu, Shu Zhang

Large language models (LLMs) have undergone significant expansion and have been increasingly integrated across various domains. Notably, in the realm of robot task planning, LLMs harness their advanced reasoning and language comprehension capabilities to formulate precise and efficient action plans based on natural language instructions. However, for embodied tasks, where robots interact with complex environments, text-only LLMs often face challenges due to a lack of compatibility with robotic visual perception. This study provides a comprehensive overview of the emerging integration of LLMs and multimodal LLMs into various robotic tasks. Additionally, we propose a framework that utilizes multimodal GPT-4V to enhance embodied task planning through the combination of natural language instructions and robot visual perceptions. Our results, based on diverse datasets, indicate that GPT-4V effectively enhances robot performance in embodied tasks. This extensive survey and evaluation of LLMs and multimodal LLMs across a variety of robotic tasks enriches the understanding of LLM-centric embodied intelligence and provides forward-looking insights toward bridging the gap in Human-Robot-Environment interaction.

http://arxiv.org/abs/2401.04334v1

cs.RO, cs.AI

Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-Shot Forecasting of Multivariate Time Series

Vijay Ekambaram, Arindam Jati, Nam H. Nguyen, Pankaj Dayama, Chandra Reddy, Wesley M. Gifford, Jayant Kalagnanam

Large pre-trained models for zero/few-shot learning excel in language and vision domains but encounter challenges in multivariate time series (TS) due to the diverse nature and scarcity of publicly available pre-training data. Consequently, there has been a recent surge in utilizing pre-trained large language models (LLMs) with token adaptations for TS forecasting. These approaches employ cross-domain transfer learning and surprisingly yield impressive results. However, these models are typically very slow and large (~billion parameters) and do not consider cross-channel correlations. To address this, we present Tiny Time Mixers (TTM), a significantly small model based on the lightweight TSMixer architecture. TTM marks the first success in developing fast and tiny general pre-trained models (<1M parameters), exclusively trained on public TS datasets, with effective transfer learning capabilities for forecasting. To tackle the complexity of pre-training on multiple datasets with varied temporal resolutions, we introduce several novel enhancements such as adaptive patching, dataset augmentation via downsampling, and resolution prefix tuning. Moreover, we employ a multi-level modeling strategy to effectively model channel correlations and infuse exogenous signals during fine-tuning, a crucial capability lacking in existing benchmarks. TTM shows significant accuracy gains (12-38\%) over popular benchmarks in few/zero-shot forecasting. It also drastically reduces the compute needs as compared to LLM-TS methods, with a 14X cut in learnable parameters, 106X less total parameters, and substantial reductions in fine-tuning (65X) and inference time (54X). In fact, TTM's zero-shot often surpasses the few-shot results in many popular benchmarks, highlighting the efficacy of our approach. Models and source code are available at https://huggingface.co/ibm/TTM

http://arxiv.org/abs/2401.03955v5

cs.LG, cs.AI

TextMachina: Seamless Generation of Machine-Generated Text Datasets

Areg Mikael Sarvazyan, José Ángel González, Marc Franco-Salvador

Recent advancements in Large Language Models (LLMs) have led to high-quality Machine-Generated Text (MGT), giving rise to countless new use cases and applications. However, easy access to LLMs is posing new challenges due to misuse. To address malicious usage, researchers have released datasets to effectively train models on MGT-related tasks. Similar strategies are used to compile these datasets, but no tool currently unifies them. In this scenario, we introduce TextMachina, a modular and extensible Python framework, designed to aid in the creation of high-quality, unbiased datasets to build robust models for MGT-related tasks such as detection, attribution, mixcase, or boundary detection. It provides a user-friendly pipeline that abstracts away the inherent intricacies of building MGT datasets, such as LLM integrations, prompt templating, and bias mitigation. The quality of the datasets generated by TextMachina has been assessed in previous works, including shared tasks where more than one hundred teams trained robust MGT detectors.

http://arxiv.org/abs/2401.03946v2

cs.CL

T-FREX: A Transformer-based Feature Extraction Method from Mobile App Reviews

Quim Motger, Alessio Miaschi, Felice Dell'Orletta, Xavier Franch, Jordi Marco

Mobile app reviews are a large-scale data source for software-related knowledge generation activities, including software maintenance, evolution and feedback analysis. Effective extraction of features (i.e., functionalities or characteristics) from these reviews is key to support analysis on the acceptance of these features, identification of relevant new feature requests and prioritization of feature development, among others. Traditional methods focus on syntactic pattern-based approaches, typically context-agnostic, evaluated on a closed set of apps, difficult to replicate and limited to a reduced set and domain of apps. Meanwhile, the pervasiveness of Large Language Models (LLMs) based on the Transformer architecture in software engineering tasks lays the groundwork for empirical evaluation of the performance of these models to support feature extraction. In this study, we present T-FREX, a Transformer-based, fully automatic approach for mobile app review feature extraction. First, we collect a set of ground truth features from users in a real crowdsourced software recommendation platform and transfer them automatically into a dataset of app reviews. Then, we use this newly created dataset to fine-tune multiple LLMs on a named entity recognition task under different data configurations. We assess the performance of T-FREX with respect to this ground truth, and we complement our analysis by comparing T-FREX with a baseline method from the field. Finally, we assess the quality of new features predicted by T-FREX through an external human evaluation. Results show that T-FREX outperforms on average the traditional syntactic-based method, especially when discovering new features from a domain for which the model has been fine-tuned.

http://arxiv.org/abs/2401.03833v1

cs.SE

Enhanced Automated Code Vulnerability Repair using Large Language Models

David de-Fitero-Dominguez, Eva Garcia-Lopez, Antonio Garcia-Cabot, Jose-Javier Martinez-Herraiz

This research addresses the complex challenge of automated repair of code vulnerabilities, vital for enhancing digital security in an increasingly technology-driven world. The study introduces a novel and efficient format for the representation of code modification, using advanced Large Language Models (LLMs) such as Code Llama and Mistral. These models, fine-tuned on datasets featuring C code vulnerabilities, significantly improve the accuracy and adaptability of automated code repair techniques. A key finding is the enhanced repair accuracy of these models when compared to previous methods such as VulRepair, which underscores their practical utility and efficiency. The research also offers a critical assessment of current evaluation metrics, such as perfect predictions, and their limitations in reflecting the true capabilities of automated repair models in real-world scenarios. Following this, it underscores the importance of using test datasets devoid of train samples, emphasizing the need for dataset integrity to enhance the effectiveness of LLMs in code repair tasks. The significance of this work is its contribution to digital security, setting new standards for automated code vulnerability repair and paving the way for future advancements in the fields of cybersecurity and artificial intelligence. The study does not only highlight the potential of LLMs in enhancing code security but also fosters further exploration and research in these crucial areas.

http://arxiv.org/abs/2401.03741v1

cs.SE, cs.CL

Language Models Understand Numbers, at Least Partially

Fangwei Zhu, Damai Dai, Zhifang Sui

Large language models (LLMs) have exhibited impressive competence in various tasks, but their opaque internal mechanisms hinder their use in mathematical problems. In this paper, we study a fundamental question: whether language models understand numbers, a basic element in math. Based on an assumption that LLMs should be capable of compressing numbers in their hidden states to solve mathematical problems, we construct a synthetic dataset comprising addition problems and utilize linear probes to read out input numbers from the hidden states. Experimental results support the existence of compressed numbers in LLMs. However, it is difficult to precisely reconstruct the original numbers, indicating that the compression process may not be lossless. Further experiments show that LLMs can utilize encoded numbers to perform arithmetic computations, and the computational ability scales up with the model size. Our preliminary research suggests that LLMs exhibit a partial understanding of numbers, offering insights for future investigations about the models' mathematical capability.

http://arxiv.org/abs/2401.03735v2

cs.CL

Assessing AI Detectors in Identifying AI-Generated Code: Implications for Education

Wei Hung Pan, Ming Jie Chok, Jonathan Leong Shan Wong, Yung Xin Shin, Yeong Shian Poon, Zhou Yang, Chun Yong Chong, David Lo, Mei Kuan Lim

Educators are increasingly concerned about the usage of Large Language Models (LLMs) such as ChatGPT in programming education, particularly regarding the potential exploitation of imperfections in Artificial Intelligence Generated Content (AIGC) Detectors for academic misconduct. In this paper, we present an empirical study where the LLM is examined for its attempts to bypass detection by AIGC Detectors. This is achieved by generating code in response to a given question using different variants. We collected a dataset comprising 5,069 samples, with each sample consisting of a textual description of a coding problem and its corresponding human-written Python solution codes. These samples were obtained from various sources, including 80 from Quescol, 3,264 from Kaggle, and 1,725 from LeetCode. From the dataset, we created 13 sets of code problem variant prompts, which were used to instruct ChatGPT to generate the outputs. Subsequently, we assessed the performance of five AIGC detectors. Our results demonstrate that existing AIGC Detectors perform poorly in distinguishing between human-written code and AI-generated code.

http://arxiv.org/abs/2401.03676v1

cs.SE, cs.AI

Using Zero-shot Prompting in the Automatic Creation and Expansion of Topic Taxonomies for Tagging Retail Banking Transactions

Daniel de S. Moraes, Pedro T. C. Santos, Polyana B. da Costa, Matheus A. S. Pinto, Ivan de J. P. Pinto, Álvaro M. G. da Veiga, Sergio Colcher, Antonio J. G. Busson, Rafael H. Rocha, Rennan Gaio, Rafael Miceli, Gabriela Tourinho, Marcos Rabaioli, Leandro Santos, Fellipe Marques, David Favaro

This work presents an unsupervised method for automatically constructing and expanding topic taxonomies using instruction-based fine-tuned LLMs (Large Language Models). We apply topic modeling and keyword extraction techniques to create initial topic taxonomies and LLMs to post-process the resulting terms and create a hierarchy. To expand an existing taxonomy with new terms, we use zero-shot prompting to find out where to add new nodes, which, to our knowledge, is the first work to present such an approach to taxonomy tasks. We use the resulting taxonomies to assign tags that characterize merchants from a retail bank dataset. To evaluate our work, we asked 12 volunteers to answer a two-part form in which we first assessed the quality of the taxonomies created and then the tags assigned to merchants based on that taxonomy. The evaluation revealed a coherence rate exceeding 90% for the chosen taxonomies. The taxonomies' expansion with LLMs also showed exciting results for parent node prediction, with an f1-score above 70% in our taxonomies.

http://arxiv.org/abs/2401.06790v2

cs.CL, cs.AI

DiarizationLM: Speaker Diarization Post-Processing with Large Language Models

Quan Wang, Yiling Huang, Guanlong Zhao, Evan Clark, Wei Xia, Hank Liao

In this paper, we introduce DiarizationLM, a framework to leverage large language models (LLM) to post-process the outputs from a speaker diarization system. Various goals can be achieved with the proposed framework, such as improving the readability of the diarized transcript, or reducing the word diarization error rate (WDER). In this framework, the outputs of the automatic speech recognition (ASR) and speaker diarization systems are represented as a compact textual format, which is included in the prompt to an optionally finetuned LLM. The outputs of the LLM can be used as the refined diarization results with the desired enhancement. As a post-processing step, this framework can be easily applied to any off-the-shelf ASR and speaker diarization systems without retraining existing components. Our experiments show that a finetuned PaLM 2-S model can reduce the WDER by rel. 55.5% on the Fisher telephone conversation dataset, and rel. 44.9% on the Callhome English dataset.

http://arxiv.org/abs/2401.03506v4

eess.AS, cs.LG, cs.SD

Expanding Horizons in HCI Research Through LLM-Driven Qualitative Analysis

Maya Grace Torii, Takahito Murakami, Yoichi Ochiai

How would research be like if we still needed to "send" papers typed with a typewriter? Our life and research environment have continually evolved, often accompanied by controversial opinions about new methodologies. In this paper, we embrace this change by introducing a new approach to qualitative analysis in HCI using Large Language Models (LLMs). We detail a method that uses LLMs for qualitative data analysis and present a quantitative framework using SBART cosine similarity for performance evaluation. Our findings indicate that LLMs not only match the efficacy of traditional analysis methods but also offer unique insights. Through a novel dataset and benchmark, we explore LLMs' characteristics in HCI research, suggesting potential avenues for further exploration and application in the field.

http://arxiv.org/abs/2401.04138v1

cs.HC, cs.AI

Exploring Large Language Model based Intelligent Agents: Definitions, Methods, and Prospects

Yuheng Cheng, Ceyao Zhang, Zhengwen Zhang, Xiangrui Meng, Sirui Hong, Wenhao Li, Zihao Wang, Zekai Wang, Feng Yin, Junhua Zhao, Xiuqiang He

Intelligent agents stand out as a potential path toward artificial general intelligence (AGI). Thus, researchers have dedicated significant effort to diverse implementations for them. Benefiting from recent progress in large language models (LLMs), LLM-based agents that use universal natural language as an interface exhibit robust generalization capabilities across various applications -- from serving as autonomous general-purpose task assistants to applications in coding, social, and economic domains, LLM-based agents offer extensive exploration opportunities. This paper surveys current research to provide an in-depth overview of LLM-based intelligent agents within single-agent and multi-agent systems. It covers their definitions, research frameworks, and foundational components such as their composition, cognitive and planning methods, tool utilization, and responses to environmental feedback. We also delve into the mechanisms of deploying LLM-based agents in multi-agent systems, including multi-role collaboration, message passing, and strategies to alleviate communication issues between agents. The discussions also shed light on popular datasets and application scenarios. We conclude by envisioning prospects for LLM-based agents, considering the evolving landscape of AI and natural language processing.

http://arxiv.org/abs/2401.03428v1

cs.AI, cs.MA

LLM-Powered Code Vulnerability Repair with Reinforcement Learning and Semantic Reward

Nafis Tanveer Islam, Joseph Khoury, Andrew Seong, Mohammad Bahrami Karkevandi, Gonzalo De La Torre Parra, Elias Bou-Harb, Peyman Najafirad

In software development, the predominant emphasis on functionality often supersedes security concerns, a trend gaining momentum with AI-driven automation tools like GitHub Copilot. These tools significantly improve developers' efficiency in functional code development. Nevertheless, it remains a notable concern that such tools are also responsible for creating insecure code, predominantly because of pre-training on publicly available repositories with vulnerable code. Moreover, developers are called the "weakest link in the chain" since they have very minimal knowledge of code security. Although existing solutions provide a reasonable solution to vulnerable code, they must adequately describe and educate the developers on code security to ensure that the security issues are not repeated. Therefore we introduce a multipurpose code vulnerability analysis system \texttt{SecRepair}, powered by a large language model, CodeGen2 assisting the developer in identifying and generating fixed code along with a complete description of the vulnerability with a code comment. Our innovative methodology uses a reinforcement learning paradigm to generate code comments augmented by a semantic reward mechanism. Inspired by how humans fix code issues, we propose an instruction-based dataset suitable for vulnerability analysis with LLMs. We further identify zero-day and N-day vulnerabilities in 6 Open Source IoT Operating Systems on GitHub. Our findings underscore that incorporating reinforcement learning coupled with semantic reward augments our model's performance, thereby fortifying its capacity to address code vulnerabilities with improved efficacy.

http://arxiv.org/abs/2401.03374v2

cs.SE, cs.AI

An Investigation of Large Language Models for Real-World Hate Speech Detection

Keyan Guo, Alexander Hu, Jaden Mu, Ziheng Shi, Ziming Zhao, Nishant Vishwamitra, Hongxin Hu

Hate speech has emerged as a major problem plaguing our social spaces today. While there have been significant efforts to address this problem, existing methods are still significantly limited in effectively detecting hate speech online. A major limitation of existing methods is that hate speech detection is a highly contextual problem, and these methods cannot fully capture the context of hate speech to make accurate predictions. Recently, large language models (LLMs) have demonstrated state-of-the-art performance in several natural language tasks. LLMs have undergone extensive training using vast amounts of natural language data, enabling them to grasp intricate contextual details. Hence, they could be used as knowledge bases for context-aware hate speech detection. However, a fundamental problem with using LLMs to detect hate speech is that there are no studies on effectively prompting LLMs for context-aware hate speech detection. In this study, we conduct a large-scale study of hate speech detection, employing five established hate speech datasets. We discover that LLMs not only match but often surpass the performance of current benchmark machine learning models in identifying hate speech. By proposing four diverse prompting strategies that optimize the use of LLMs in detecting hate speech. Our study reveals that a meticulously crafted reasoning prompt can effectively capture the context of hate speech by fully utilizing the knowledge base in LLMs, significantly outperforming existing techniques. Furthermore, although LLMs can provide a rich knowledge base for the contextual detection of hate speech, suitable prompting strategies play a crucial role in effectively leveraging this knowledge base for efficient detection.

http://arxiv.org/abs/2401.03346v1

cs.CY, cs.AI, cs.CL, cs.LG, cs.SI

Reflections on Inductive Thematic Saturation as a potential metric for measuring the validity of an inductive Thematic Analysis with LLMs

Stefano De Paoli, Walter Stan Mathis

This paper presents a set of reflections on saturation and the use of Large Language Models (LLMs) for performing Thematic Analysis (TA). The paper suggests that initial thematic saturation (ITS) could be used as a metric to assess part of the transactional validity of TA with LLM, focusing on the initial coding. The paper presents the initial coding of two datasets of different sizes, and it reflects on how the LLM reaches some form of analytical saturation during the coding. The procedure proposed in this work leads to the creation of two codebooks, one comprising the total cumulative initial codes and the other the total unique codes. The paper proposes a metric to synthetically measure ITS using a simple mathematical calculation employing the ratio between slopes of cumulative codes and unique codes. The paper contributes to the initial body of work exploring how to perform qualitative analysis with LLMs.

http://arxiv.org/abs/2401.03239v1

cs.CL, cs.CY

Using Large Language Models to Assess Tutors' Performance in Reacting to Students Making Math Errors

Sanjit Kakarla, Danielle Thomas, Jionghao Lin, Shivang Gupta, Kenneth R. Koedinger

Research suggests that tutors should adopt a strategic approach when addressing math errors made by low-efficacy students. Rather than drawing direct attention to the error, tutors should guide the students to identify and correct their mistakes on their own. While tutor lessons have introduced this pedagogical skill, human evaluation of tutors applying this strategy is arduous and time-consuming. Large language models (LLMs) show promise in providing real-time assessment to tutors during their actual tutoring sessions, yet little is known regarding their accuracy in this context. In this study, we investigate the capacity of generative AI to evaluate real-life tutors' performance in responding to students making math errors. By analyzing 50 real-life tutoring dialogues, we find both GPT-3.5-Turbo and GPT-4 demonstrate proficiency in assessing the criteria related to reacting to students making errors. However, both models exhibit limitations in recognizing instances where the student made an error. Notably, GPT-4 tends to overidentify instances of students making errors, often attributing student uncertainty or inferring potential errors where human evaluators did not. Future work will focus on enhancing generalizability by assessing a larger dataset of dialogues and evaluating learning transfer. Specifically, we will analyze the performance of tutors in real-life scenarios when responding to students' math errors before and after lesson completion on this crucial tutoring skill.

http://arxiv.org/abs/2401.03238v1

cs.HC, cs.AI, cs.CY

3DMIT: 3D Multi-modal Instruction Tuning for Scene Understanding

Zeju Li, Chao Zhang, Xiaoyan Wang, Ruilong Ren, Yifan Xu, Ruifei Ma, Xiangde Liu

The remarkable potential of multi-modal large language models (MLLMs) in comprehending both vision and language information has been widely acknowledged. However, the scarcity of 3D scenes-language pairs in comparison to their 2D counterparts, coupled with the inadequacy of existing approaches in understanding of 3D scenes by LLMs, poses a significant challenge. In response, we collect and construct an extensive dataset comprising 75K instruction-response pairs tailored for 3D scenes. This dataset addresses tasks related to 3D VQA, 3D grounding, and 3D conversation. To further enhance the integration of 3D spatial information into LLMs, we introduce a novel and efficient prompt tuning paradigm, 3DMIT. This paradigm eliminates the alignment stage between 3D scenes and language and extends the instruction prompt with the 3D modality information including the entire scene and segmented objects. We evaluate the effectiveness of our method across diverse tasks in the 3D scene domain and find that our approach serves as a strategic means to enrich LLMs' comprehension of the 3D world. Our code is available at https://github.com/staymylove/3DMIT.

http://arxiv.org/abs/2401.03201v2

cs.CV, cs.MM

δ-CAUSAL: Exploring Defeasibility in Causal Reasoning

Shaobo Cui, Lazar Milikic, Yiyang Feng, Mete Ismayilzada, Debjit Paul, Antoine Bosselut, Boi Faltings

Defeasibility in causal reasoning implies that the causal relationship between cause and effect can be strengthened or weakened. Namely, the causal strength between cause and effect should increase or decrease with the incorporation of strengthening arguments (supporters) or weakening arguments (defeaters), respectively. However, existing works ignore defeasibility in causal reasoning and fail to evaluate existing causal strength metrics in defeasible settings. In this work, we present {\delta}-CAUSAL, the first benchmark dataset for studying defeasibility in causal reasoning. {\delta}-CAUSAL includes around 11K events spanning ten domains, featuring defeasible causality pairs, i.e., cause-effect pairs accompanied by supporters and defeaters. We further show current causal strength metrics fail to reflect the change of causal strength with the incorporation of supporters or defeaters in {\delta}-CAUSAL. To this end, we propose CESAR (Causal Embedding aSsociation with Attention Rating), a metric that measures causal strength based on token-level causal relationships. CESAR achieves a significant 69.7% relative improvement over existing metrics, increasing from 47.2% to 80.1% in capturing the causal strength change brought by supporters and defeaters. We further demonstrate even Large Language Models (LLMs) like GPT-3.5 still lag 4.5 and 10.7 points behind humans in generating supporters and defeaters, emphasizing the challenge posed by {\delta}-CAUSAL.

http://arxiv.org/abs/2401.03183v1

cs.CL

Quartet Logic: A Four-Step Reasoning (QLFR) framework for advancing Short Text Classification

Hui Wu, Yuanben Zhang, Zhonghe Han, Yingyan Hou, Lei Wang, Siye Liu, Qihang Gong, Yunping Ge

Short Text Classification (STC) is crucial for processing and comprehending the brief but substantial content prevalent on contemporary digital platforms. The STC encounters difficulties in grasping semantic and syntactic intricacies, an issue that is apparent in traditional pre-trained language models. Although Graph Convolutional Networks enhance performance by integrating external knowledge bases, these methods are limited by the quality and extent of the knowledge applied. Recently, the emergence of Large Language Models (LLMs) and Chain-of-Thought (CoT) has significantly improved the performance of complex reasoning tasks. However, some studies have highlighted the limitations of their application in fundamental NLP tasks. Consequently, this study sought to employ CoT to investigate the capabilities of LLMs in STC tasks. This study introduces Quartet Logic: A Four-Step Reasoning (QLFR) framework. This framework primarily incorporates Syntactic and Semantic Enrichment CoT, effectively decomposing the STC task into four distinct steps: (i) essential concept identification, (ii) common-sense knowledge retrieval, (iii) text rewriting, and (iv) classification. This elicits the inherent knowledge and abilities of LLMs to address the challenges in STC. Surprisingly, we found that QLFR can also improve the performance of smaller models. Therefore, we developed a CoT-Driven Multi-task learning (QLFR-CML) method to facilitate the knowledge transfer from LLMs to smaller models. Extensive experimentation across six short-text benchmarks validated the efficacy of the proposed methods. Notably, QLFR achieved state-of-the-art performance on all datasets, with significant improvements, particularly on the Ohsumed and TagMyNews datasets.

http://arxiv.org/abs/2401.03158v1

cs.CL, cs.AI

DeepSeek LLM: Scaling Open-Source Language Models with Longtermism

DeepSeek-AI, :, Xiao Bi, Deli Chen, Guanting Chen, Shanhuang Chen, Damai Dai, Chengqi Deng, Honghui Ding, Kai Dong, Qiushi Du, Zhe Fu, Huazuo Gao, Kaige Gao, Wenjun Gao, Ruiqi Ge, Kang Guan, Daya Guo, Jianzhong Guo, Guangbo Hao, Zhewen Hao, Ying He, Wenjie Hu, Panpan Huang, Erhang Li, Guowei Li, Jiashi Li, Yao Li, Y. K. Li, Wenfeng Liang, Fangyun Lin, A. X. Liu, Bo Liu, Wen Liu, Xiaodong Liu, Xin Liu, Yiyuan Liu, Haoyu Lu, Shanghao Lu, Fuli Luo, Shirong Ma, Xiaotao Nie, Tian Pei, Yishi Piao, Junjie Qiu, Hui Qu, Tongzheng Ren, Zehui Ren, Chong Ruan, Zhangli Sha, Zhihong Shao, Junxiao Song, Xuecheng Su, Jingxiang Sun, Yaofeng Sun, Minghui Tang, Bingxuan Wang, Peiyi Wang, Shiyu Wang, Yaohui Wang, Yongji Wang, Tong Wu, Y. Wu, Xin Xie, Zhenda Xie, Ziwei Xie, Yiliang Xiong, Hanwei Xu, R. X. Xu, Yanhong Xu, Dejian Yang, Yuxiang You, Shuiping Yu, Xingkai Yu, B. Zhang, Haowei Zhang, Lecong Zhang, Liyue Zhang, Mingchuan Zhang, Minghua Zhang, Wentao Zhang, Yichao Zhang, Chenggang Zhao, Yao Zhao, Shangyan Zhou, Shunfeng Zhou, Qihao Zhu, Yuheng Zou

The rapid development of open-source large language models (LLMs) has been truly remarkable. However, the scaling law described in previous literature presents varying conclusions, which casts a dark cloud over scaling LLMs. We delve into the study of scaling laws and present our distinctive findings that facilitate scaling of large scale models in two commonly used open-source configurations, 7B and 67B. Guided by the scaling laws, we introduce DeepSeek LLM, a project dedicated to advancing open-source language models with a long-term perspective. To support the pre-training phase, we have developed a dataset that currently consists of 2 trillion tokens and is continuously expanding. We further conduct supervised fine-tuning (SFT) and Direct Preference Optimization (DPO) on DeepSeek LLM Base models, resulting in the creation of DeepSeek Chat models. Our evaluation results demonstrate that DeepSeek LLM 67B surpasses LLaMA-2 70B on various benchmarks, particularly in the domains of code, mathematics, and reasoning. Furthermore, open-ended evaluations reveal that DeepSeek LLM 67B Chat exhibits superior performance compared to GPT-3.5.

http://arxiv.org/abs/2401.02954v1

cs.CL, cs.AI, cs.LG

Introducing Bode: A Fine-Tuned Large Language Model for Portuguese Prompt-Based Task

Gabriel Lino Garcia, Pedro Henrique Paiola, Luis Henrique Morelli, Giovani Candido, Arnaldo Cândido Júnior, Danilo Samuel Jodas, Luis C. S. Afonso, Ivan Rizzo Guilherme, Bruno Elias Penteado, João Paulo Papa

Large Language Models (LLMs) are increasingly bringing advances to Natural Language Processing. However, low-resource languages, those lacking extensive prominence in datasets for various NLP tasks, or where existing datasets are not as substantial, such as Portuguese, already obtain several benefits from LLMs, but not to the same extent. LLMs trained on multilingual datasets normally struggle to respond to prompts in Portuguese satisfactorily, presenting, for example, code switching in their responses. This work proposes a fine-tuned LLaMA 2-based model for Portuguese prompts named Bode in two versions: 7B and 13B. We evaluate the performance of this model in classification tasks using the zero-shot approach with in-context learning, and compare it with other LLMs. Our main contribution is to bring an LLM with satisfactory results in the Portuguese language, as well as to provide a model that is free for research or commercial purposes.

http://arxiv.org/abs/2401.02909v1

cs.CL

DocGraphLM: Documental Graph Language Model for Information Extraction

Dongsheng Wang, Zhiqiang Ma, Armineh Nourbakhsh, Kang Gu, Sameena Shah

Advances in Visually Rich Document Understanding (VrDU) have enabled information extraction and question answering over documents with complex layouts. Two tropes of architectures have emerged -- transformer-based models inspired by LLMs, and Graph Neural Networks. In this paper, we introduce DocGraphLM, a novel framework that combines pre-trained language models with graph semantics. To achieve this, we propose 1) a joint encoder architecture to represent documents, and 2) a novel link prediction approach to reconstruct document graphs. DocGraphLM predicts both directions and distances between nodes using a convergent joint loss function that prioritizes neighborhood restoration and downweighs distant node detection. Our experiments on three SotA datasets show consistent improvement on IE and QA tasks with the adoption of graph features. Moreover, we report that adopting the graph features accelerates convergence in the learning process during training, despite being solely constructed through link prediction.

http://arxiv.org/abs/2401.02823v1

cs.CL, cs.IR

Infinite-LLM: Efficient LLM Service for Long Context with DistAttention and Distributed KVCache

Bin Lin, Tao Peng, Chen Zhang, Minmin Sun, Lanbo Li, Hanyu Zhao, Wencong Xiao, Qi Xu, Xiafei Qiu, Shen Li, Zhigang Ji, Yong Li, Wei Lin

The rapid proliferation of Large Language Models (LLMs) has been a driving force in the growth of cloud-based LLM services, which are now integral to advancing AI applications. However, the dynamic auto-regressive nature of LLM service, along with the need to support exceptionally long context lengths, demands the flexible allocation and release of substantial resources. This presents considerable challenges in designing cloud-based LLM service systems, where inefficient management can lead to performance degradation or resource wastage. In response to these challenges, this work introduces DistAttention, a novel distributed attention algorithm that segments the KV Cache into smaller, manageable units, enabling distributed processing and storage of the attention module. Based on that, we propose DistKV-LLM, a distributed LLM serving system that dynamically manages KV Cache and effectively orchestrates all accessible GPU and CPU memories spanning across the data center. This ensures a high-performance LLM service on the cloud, adaptable to a broad range of context lengths. Validated in a cloud environment with 32 NVIDIA A100 GPUs in configurations from 2 to 32 instances, our system exhibited 1.03-2.4x end-to-end throughput improvements and supported context lengths 2-19x longer than current state-of-the-art LLM service systems, as evidenced by extensive testing across 18 datasets with context lengths up to 1,900K.

http://arxiv.org/abs/2401.02669v1

cs.DC, cs.AR

Learning to Prompt with Text Only Supervision for Vision-Language Models

Muhammad Uzair Khattak, Muhammad Ferjad Naeem, Muzammal Naseer, Luc Van Gool, Federico Tombari

Foundational vision-language models such as CLIP are becoming a new paradigm in vision, due to their excellent generalization abilities. However, adapting these models for downstream tasks while maintaining their generalization remains a challenge. In literature, one branch of methods adapts CLIP by learning prompts using visual information. While effective, most of these works require labeled data which is not practical, and often struggle to generalize towards new datasets due to over-fitting on the source data. An alternative approach resorts to training-free methods by generating class descriptions from large language models (LLMs) and perform prompt ensembling. However, these methods often generate class specific prompts that cannot be transferred to other classes, which incur higher costs by generating LLM descriptions for each class separately. In this work, we propose to combine the strengths of these both streams of methods by learning prompts using only text data derived from LLMs. As supervised training of prompts is not trivial due to absence of images, we develop a training approach that allows prompts to extract rich contextual knowledge from LLM data. Moreover, with LLM contextual data mapped within the learned prompts, it enables zero-shot transfer of prompts to new classes and datasets potentially cutting the LLM prompt engineering cost. To the best of our knowledge, this is the first work that learns generalized prompts using text only data. We perform extensive evaluations on 4 benchmarks where our method improves over prior ensembling works while being competitive to those utilizing labeled images. Our code and pre-trained models are available at https://github.com/muzairkhattak/ProText.

http://arxiv.org/abs/2401.02418v1

cs.CV

SPEER: Sentence-Level Planning of Long Clinical Summaries via Embedded Entity Retrieval

Griffin Adams, Jason Zucker, Noémie Elhadad

Clinician must write a lengthy summary each time a patient is discharged from the hospital. This task is time-consuming due to the sheer number of unique clinical concepts covered in the admission. Identifying and covering salient entities is vital for the summary to be clinically useful. We fine-tune open-source LLMs (Mistral-7B-Instruct and Zephyr-7B-\b{eta}) on the task and find that they generate incomplete and unfaithful summaries. To increase entity coverage, we train a smaller, encoder-only model to predict salient entities, which are treated as content-plans to guide the LLM. To encourage the LLM to focus on specific mentions in the source notes, we propose SPEER: Sentence-level Planning via Embedded Entity Retrieval. Specifically, we mark each salient entity span with special "{{ }}" boundary tags and instruct the LLM to retrieve marked spans before generating each sentence. Sentence-level planning acts as a form of state tracking in that the model is explicitly recording the entities it uses. We fine-tune Mistral and Zephyr variants on a large-scale, diverse dataset of ~167k in-patient hospital admissions and evaluate on 3 datasets. SPEER shows gains in both coverage and faithfulness metrics over non-guided and guided baselines.

http://arxiv.org/abs/2401.02369v1

cs.CL

Are LLMs Robust for Spoken Dialogues?

Seyed Mahed Mousavi, Gabriel Roccabruna, Simone Alghisi, Massimo Rizzoli, Mirco Ravanelli, Giuseppe Riccardi

Large Pre-Trained Language Models have demonstrated state-of-the-art performance in different downstream tasks, including dialogue state tracking and end-to-end response generation. Nevertheless, most of the publicly available datasets and benchmarks on task-oriented dialogues focus on written conversations. Consequently, the robustness of the developed models to spoken interactions is unknown. In this work, we have evaluated the performance of LLMs for spoken task-oriented dialogues on the DSTC11 test sets. Due to the lack of proper spoken dialogue datasets, we have automatically transcribed a development set of spoken dialogues with a state-of-the-art ASR engine. We have characterized the ASR-error types and their distributions and simulated these errors in a large dataset of dialogues. We report the intrinsic (perplexity) and extrinsic (human evaluation) performance of fine-tuned GPT-2 and T5 models in two subtasks of response generation and dialogue state tracking, respectively. The results show that LLMs are not robust to spoken noise by default, however, fine-tuning/training such models on a proper dataset of spoken TODs can result in a more robust performance.

http://arxiv.org/abs/2401.02297v1

cs.CL

A case study of Generative AI in MSX Sales Copilot: Improving seller productivity with a real-time question-answering system for content recommendation

Manpreet Singh, Ravdeep Pasricha, Nitish Singh, Ravi Prasad Kondapalli, Manoj R, Kiran R, Laurent Boué

In this paper, we design a real-time question-answering system specifically targeted for helping sellers get relevant material/documentation they can share live with their customers or refer to during a call. Taking the Seismic content repository as a relatively large scale example of a diverse dataset of sales material, we demonstrate how LLM embeddings of sellers' queries can be matched with the relevant content. We achieve this by engineering prompts in an elaborate fashion that makes use of the rich set of meta-features available for documents and sellers. Using a bi-encoder with cross-encoder re-ranker architecture, we show how the solution returns the most relevant content recommendations in just a few seconds even for large datasets. Our recommender system is deployed as an AML endpoint for real-time inferencing and has been integrated into a Copilot interface that is now deployed in the production version of the Dynamics CRM, known as MSX, used daily by Microsoft sellers.

http://arxiv.org/abs/2401.04732v1

cs.IR, cs.AI, cs.LG

Improved Zero-Shot Classification by Adapting VLMs with Text Descriptions

Oindrila Saha, Grant Van Horn, Subhransu Maji

The zero-shot performance of existing vision-language models (VLMs) such as CLIP is limited by the availability of large-scale, aligned image and text datasets in specific domains. In this work, we leverage two complementary sources of information -- descriptions of categories generated by large language models (LLMs) and abundant, fine-grained image classification datasets -- to improve the zero-shot classification performance of VLMs across fine-grained domains. On the technical side, we develop methods to train VLMs with this "bag-level" image-text supervision. We find that simply using these attributes at test-time does not improve performance, but our training strategy, for example, on the iNaturalist dataset, leads to an average improvement of 4-5% in zero-shot classification accuracy for novel categories of birds and flowers. Similar improvements are observed in domains where a subset of the categories was used to fine-tune the model. By prompting LLMs in various ways, we generate descriptions that capture visual appearance, habitat, and geographic regions and pair them with existing attributes such as the taxonomic structure of the categories. We systematically evaluate their ability to improve zero-shot categorization in natural domains. Our findings suggest that geographic priors can be just as effective and are complementary to visual appearance. Our method also outperforms prior work on prompt-based tuning of VLMs. We release the benchmark, consisting of 14 datasets at https://github.com/cvl-umass/AdaptCLIPZS , which will contribute to future research in zero-shot recognition.

http://arxiv.org/abs/2401.02460v2

cs.CV

DCR-Consistency: Divide-Conquer-Reasoning for Consistency Evaluation and Improvement of Large Language Models

Wendi Cui, Jiaxin Zhang, Zhuohang Li, Lopez Damien, Kamalika Das, Bradley Malin, Sricharan Kumar

Evaluating the quality and variability of text generated by Large Language Models (LLMs) poses a significant, yet unresolved research challenge. Traditional evaluation methods, such as ROUGE and BERTScore, which measure token similarity, often fail to capture the holistic semantic equivalence. This results in a low correlation with human judgments and intuition, which is especially problematic in high-stakes applications like healthcare and finance where reliability, safety, and robust decision-making are highly critical. This work proposes DCR, an automated framework for evaluating and improving the consistency of LLM-generated texts using a divide-conquer-reasoning approach. Unlike existing LLM-based evaluators that operate at the paragraph level, our method employs a divide-and-conquer evaluator (DCE) that breaks down the paragraph-to-paragraph comparison between two generated responses into individual sentence-to-paragraph comparisons, each evaluated based on predefined criteria. To facilitate this approach, we introduce an automatic metric converter (AMC) that translates the output from DCE into an interpretable numeric score. Beyond the consistency evaluation, we further present a reason-assisted improver (RAI) that leverages the analytical reasons with explanations identified by DCE to generate new responses aimed at reducing these inconsistencies. Through comprehensive and systematic empirical analysis, we show that our approach outperforms state-of-the-art methods by a large margin (e.g., +19.3% and +24.3% on the SummEval dataset) in evaluating the consistency of LLM generation across multiple benchmarks in semantic, factual, and summarization consistency tasks. Our approach also substantially reduces nearly 90% of output inconsistencies, showing promise for effective hallucination mitigation.

http://arxiv.org/abs/2401.02132v1

cs.CL, cs.AI

Using LLM to select the right SQL Query from candidates

Zhenwen Li, Tao Xie

Text-to-SQL models can generate a list of candidate SQL queries, and the best query is often in the candidate list, but not at the top of the list. An effective re-rank method can select the right SQL query from the candidate list and improve the model's performance. Previous studies on code generation automatically generate test cases and use them to re-rank candidate codes. However, automatic test case generation for text-to-SQL is an understudied field. We propose an automatic test case generation method that first generates a database and then uses LLMs to predict the ground truth, which is the expected execution results of the ground truth SQL query on this database. To reduce the difficulty for LLMs to predict, we conduct experiments to search for ways to generate easy databases for LLMs and design easy-to-understand prompts. Based on our test case generation method, we propose a re-rank method to select the right SQL query from the candidate list. Given a candidate list, our method can generate test cases and re-rank the candidate list according to their pass numbers on these test cases and their generation probabilities. The experiment results on the validation dataset of Spider show that the performance of some state-of-the-art models can get a 3.6\% improvement after applying our re-rank method.

http://arxiv.org/abs/2401.02115v1

cs.CL

Text2MDT: Extracting Medical Decision Trees from Medical Texts

Wei Zhu, Wenfeng Li, Xing Tian, Pengfei Wang, Xiaoling Wang, Jin Chen, Yuanbin Wu, Yuan Ni, Guotong Xie

Knowledge of the medical decision process, which can be modeled as medical decision trees (MDTs), is critical to build clinical decision support systems. However, the current MDT construction methods rely heavily on time-consuming and laborious manual annotation. In this work, we propose a novel task, Text2MDT, to explore the automatic extraction of MDTs from medical texts such as medical guidelines and textbooks. We normalize the form of the MDT and create an annotated Text-to-MDT dataset in Chinese with the participation of medical experts. We investigate two different methods for the Text2MDT tasks: (a) an end-to-end framework which only relies on a GPT style large language models (LLM) instruction tuning to generate all the node information and tree structures. (b) The pipeline framework which decomposes the Text2MDT task to three subtasks. Experiments on our Text2MDT dataset demonstrate that: (a) the end-to-end method basd on LLMs (7B parameters or larger) show promising results, and successfully outperform the pipeline methods. (b) The chain-of-thought (COT) prompting method \cite{Wei2022ChainOT} can improve the performance of the fine-tuned LLMs on the Text2MDT test set. (c) the lightweight pipelined method based on encoder-based pretrained models can perform comparably with LLMs with model complexity two magnititudes smaller. Our Text2MDT dataset is open-sourced at \url{https://tianchi.aliyun.com/dataset/95414}, and the source codes are open-sourced at \url{https://github.com/michael-wzhu/text2dt}.

http://arxiv.org/abs/2401.02034v1

cs.CL

Revisiting Zero-Shot Abstractive Summarization in the Era of Large Language Models from the Perspective of Position Bias

Anshuman Chhabra, Hadi Askari, Prasant Mohapatra

We characterize and study zero-shot abstractive summarization in Large Language Models (LLMs) by measuring position bias, which we propose as a general formulation of the more restrictive lead bias phenomenon studied previously in the literature. Position bias captures the tendency of a model unfairly prioritizing information from certain parts of the input text over others, leading to undesirable behavior. Through numerous experiments on four diverse real-world datasets, we study position bias in multiple LLM models such as GPT 3.5-Turbo, Llama-2, and Dolly-v2, as well as state-of-the-art pretrained encoder-decoder abstractive summarization models such as Pegasus and BART. Our findings lead to novel insights and discussion on performance and position bias of models for zero-shot summarization tasks.

http://arxiv.org/abs/2401.01989v3

cs.CL, cs.AI

Generalist embedding models are better at short-context clinical semantic search than specialized embedding models

Jean-Baptiste Excoffier, Tom Roehr, Alexei Figueroa, Jens-Michalis Papaioannou, Keno Bressem, Matthieu Ortala

The increasing use of tools and solutions based on Large Language Models (LLMs) for various tasks in the medical domain has become a prominent trend. Their use in this highly critical and sensitive domain has thus raised important questions about their robustness, especially in response to variations in input, and the reliability of the generated outputs. This study addresses these questions by constructing a textual dataset based on the ICD-10-CM code descriptions, widely used in US hospitals and containing many clinical terms, and their easily reproducible rephrasing. We then benchmarked existing embedding models, either generalist or specialized in the clinical domain, in a semantic search task where the goal was to correctly match the rephrased text to the original description. Our results showed that generalist models performed better than clinical models, suggesting that existing clinical specialized models are more sensitive to small changes in input that confuse them. The highlighted problem of specialized models may be due to the fact that they have not been trained on sufficient data, and in particular on datasets that are not diverse enough to have a reliable global language understanding, which is still necessary for accurate handling of medical documents.

http://arxiv.org/abs/2401.01943v2

cs.CL, cs.AI

De-Hallucinator: Iterative Grounding for LLM-Based Code Completion

Aryaz Eghbali, Michael Pradel

Large languages models (LLMs) trained on datasets of publicly available source code have established a new state-of-the-art in code completion. However, these models are mostly unaware of the code that already exists within a specific project, preventing the models from making good use of existing APIs. Instead, LLMs often invent, or "hallucinate", non-existent APIs or produce variants of already existing code. Although the API information is available to IDEs, the input size limit of LLMs prevents code completion techniques from including all relevant context into the prompt. This paper presents De-Hallucinator, an LLM-based code completion technique that grounds the predictions of a model through a novel combination of retrieving suitable API references and iteratively querying the model with increasingly suitable context information in the prompt. The approach exploits the observation that LLMs often predict code that resembles the desired completion, but that fails to correctly refer to already existing APIs. De-Hallucinator automatically identifies project-specific API references related to the code prefix and to the model's initial predictions and adds these references into the prompt. Our evaluation applies the approach to the task of predicting API usages in open-source Python projects. We show that De-Hallucinator consistently improves the predicted code across four state-of-the-art LLMs compared to querying the model only with the code before the cursor. In particular, the approach improves the edit distance of the predicted code by 23-51% and the recall of correctly predicted API usages by 24-61% relative to the baseline.

http://arxiv.org/abs/2401.01701v2

cs.SE

PLLaMa: An Open-source Large Language Model for Plant Science

Xianjun Yang, Junfeng Gao, Wenxin Xue, Erik Alexandersson

Large Language Models (LLMs) have exhibited remarkable capabilities in understanding and interacting with natural language across various sectors. However, their effectiveness is limited in specialized areas requiring high accuracy, such as plant science, due to a lack of specific expertise in these fields. This paper introduces PLLaMa, an open-source language model that evolved from LLaMa-2. It's enhanced with a comprehensive database, comprising more than 1.5 million scholarly articles in plant science. This development significantly enriches PLLaMa with extensive knowledge and proficiency in plant and agricultural sciences. Our initial tests, involving specific datasets related to plants and agriculture, show that PLLaMa substantially improves its understanding of plant science-related topics. Moreover, we have formed an international panel of professionals, including plant scientists, agricultural engineers, and plant breeders. This team plays a crucial role in verifying the accuracy of PLLaMa's responses to various academic inquiries, ensuring its effective and reliable application in the field. To support further research and development, we have made the model's checkpoints and source codes accessible to the scientific community. These resources are available for download at \url{https://github.com/Xianjun-Yang/PLLaMa}.

http://arxiv.org/abs/2401.01600v1

cs.CL, cs.AI, cs.CE, cs.LG

MedSumm: A Multimodal Approach to Summarizing Code-Mixed Hindi-English Clinical Queries

Akash Ghosh, Arkadeep Acharya, Prince Jha, Aniket Gaudgaul, Rajdeep Majumdar, Sriparna Saha, Aman Chadha, Raghav Jain, Setu Sinha, Shivani Agarwal

In the healthcare domain, summarizing medical questions posed by patients is critical for improving doctor-patient interactions and medical decision-making. Although medical data has grown in complexity and quantity, the current body of research in this domain has primarily concentrated on text-based methods, overlooking the integration of visual cues. Also prior works in the area of medical question summarisation have been limited to the English language. This work introduces the task of multimodal medical question summarization for codemixed input in a low-resource setting. To address this gap, we introduce the Multimodal Medical Codemixed Question Summarization MMCQS dataset, which combines Hindi-English codemixed medical queries with visual aids. This integration enriches the representation of a patient's medical condition, providing a more comprehensive perspective. We also propose a framework named MedSumm that leverages the power of LLMs and VLMs for this task. By utilizing our MMCQS dataset, we demonstrate the value of integrating visual information from images to improve the creation of medically detailed summaries. This multimodal strategy not only improves healthcare decision-making but also promotes a deeper comprehension of patient queries, paving the way for future exploration in personalized and responsive medical care. Our dataset, code, and pre-trained models will be made publicly available.

http://arxiv.org/abs/2401.01596v1

cs.AI, cs.CL

AstroLLaMA-Chat: Scaling AstroLLaMA with Conversational and Diverse Datasets

Ernest Perkowski, Rui Pan, Tuan Dung Nguyen, Yuan-Sen Ting, Sandor Kruk, Tong Zhang, Charlie O'Neill, Maja Jablonska, Zechang Sun, Michael J. Smith, Huiling Liu, Kevin Schawinski, Kartheik Iyer, Ioana Ciucă for UniverseTBD

We explore the potential of enhancing LLM performance in astronomy-focused question-answering through targeted, continual pre-training. By employing a compact 7B-parameter LLaMA-2 model and focusing exclusively on a curated set of astronomy corpora -- comprising abstracts, introductions, and conclusions -- we achieve notable improvements in specialized topic comprehension. While general LLMs like GPT-4 excel in broader question-answering scenarios due to superior reasoning capabilities, our findings suggest that continual pre-training with limited resources can still enhance model performance on specialized topics. Additionally, we present an extension of AstroLLaMA: the fine-tuning of the 7B LLaMA model on a domain-specific conversational dataset, culminating in the release of the chat-enabled AstroLLaMA for community use. Comprehensive quantitative benchmarking is currently in progress and will be detailed in an upcoming full paper. The model, AstroLLaMA-Chat, is now available at https://huggingface.co/universeTBD, providing the first open-source conversational AI tool tailored for the astronomy community.

http://arxiv.org/abs/2401.01916v2

astro-ph.IM, astro-ph.CO, astro-ph.GA, astro-ph.SR, cs.CL, cs.LG

Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models

Zixiang Chen, Yihe Deng, Huizhuo Yuan, Kaixuan Ji, Quanquan Gu

Harnessing the power of human-annotated data through Supervised Fine-Tuning (SFT) is pivotal for advancing Large Language Models (LLMs). In this paper, we delve into the prospect of growing a strong LLM out of a weak one without the need for acquiring additional human-annotated data. We propose a new fine-tuning method called Self-Play fIne-tuNing (SPIN), which starts from a supervised fine-tuned model. At the heart of SPIN lies a self-play mechanism, where the LLM refines its capability by playing against instances of itself. More specifically, the LLM generates its own training data from its previous iterations, refining its policy by discerning these self-generated responses from those obtained from human-annotated data. Our method progressively elevates the LLM from a nascent model to a formidable one, unlocking the full potential of human-annotated demonstration data for SFT. Theoretically, we prove that the global optimum to the training objective function of our method is achieved only when the LLM policy aligns with the target data distribution. Empirically, we evaluate our method on several benchmark datasets including the HuggingFace Open LLM Leaderboard, MT-Bench, and datasets from Big-Bench. Our results show that SPIN can significantly improve the LLM's performance across a variety of benchmarks and even outperform models trained through direct preference optimization (DPO) supplemented with extra GPT-4 preference data. This sheds light on the promise of self-play, enabling the achievement of human-level performance in LLMs without the need for expert opponents. Codes are available at https://github.com/uclaml/SPIN.

http://arxiv.org/abs/2401.01335v2

cs.LG, cs.AI, cs.CL, stat.ML

A Comprehensive Survey of Hallucination Mitigation Techniques in Large Language Models

S. M Towhidul Islam Tonmoy, S M Mehedi Zaman, Vinija Jain, Anku Rani, Vipula Rawte, Aman Chadha, Amitava Das

As Large Language Models (LLMs) continue to advance in their ability to write human-like text, a key challenge remains around their tendency to hallucinate generating content that appears factual but is ungrounded. This issue of hallucination is arguably the biggest hindrance to safely deploying these powerful LLMs into real-world production systems that impact people's lives. The journey toward widespread adoption of LLMs in practical settings heavily relies on addressing and mitigating hallucinations. Unlike traditional AI systems focused on limited tasks, LLMs have been exposed to vast amounts of online text data during training. While this allows them to display impressive language fluency, it also means they are capable of extrapolating information from the biases in training data, misinterpreting ambiguous prompts, or modifying the information to align superficially with the input. This becomes hugely alarming when we rely on language generation capabilities for sensitive applications, such as summarizing medical records, financial analysis reports, etc. This paper presents a comprehensive survey of over 32 techniques developed to mitigate hallucination in LLMs. Notable among these are Retrieval Augmented Generation (Lewis et al, 2021), Knowledge Retrieval (Varshney et al,2023), CoNLI (Lei et al, 2023), and CoVe (Dhuliawala et al, 2023). Furthermore, we introduce a detailed taxonomy categorizing these methods based on various parameters, such as dataset utilization, common tasks, feedback mechanisms, and retriever types. This classification helps distinguish the diverse approaches specifically designed to tackle hallucination issues in LLMs. Additionally, we analyze the challenges and limitations inherent in these techniques, providing a solid foundation for future research in addressing hallucinations and related phenomena within the realm of LLMs.

http://arxiv.org/abs/2401.01313v3

cs.CL

A Comprehensive Study of Knowledge Editing for Large Language Models

Ningyu Zhang, Yunzhi Yao, Bozhong Tian, Peng Wang, Shumin Deng, Mengru Wang, Zekun Xi, Shengyu Mao, Jintian Zhang, Yuansheng Ni, Siyuan Cheng, Ziwen Xu, Xin Xu, Jia-Chen Gu, Yong Jiang, Pengjun Xie, Fei Huang, Lei Liang, Zhiqiang Zhang, Xiaowei Zhu, Jun Zhou, Huajun Chen

Large Language Models (LLMs) have shown extraordinary capabilities in understanding and generating text that closely mirrors human communication. However, a primary limitation lies in the significant computational demands during training, arising from their extensive parameterization. This challenge is further intensified by the dynamic nature of the world, necessitating frequent updates to LLMs to correct outdated information or integrate new knowledge, thereby ensuring their continued relevance. Note that many applications demand continual model adjustments post-training to address deficiencies or undesirable behaviors. There is an increasing interest in efficient, lightweight methods for on-the-fly model modifications. To this end, recent years have seen a burgeoning in the techniques of knowledge editing for LLMs, which aim to efficiently modify LLMs' behaviors within specific domains while preserving overall performance across various inputs. In this paper, we first define the knowledge editing problem and then provide a comprehensive review of cutting-edge approaches. Drawing inspiration from educational and cognitive research theories, we propose a unified categorization criterion that classifies knowledge editing methods into three groups: resorting to external knowledge, merging knowledge into the model, and editing intrinsic knowledge. Furthermore, we introduce a new benchmark, KnowEdit, for a comprehensive empirical evaluation of representative knowledge editing approaches. Additionally, we provide an in-depth analysis of knowledge location, which can give a deeper understanding of the knowledge structures inherent within LLMs. Finally, we discuss several potential applications of knowledge editing, outlining its broad and impactful implications.

http://arxiv.org/abs/2401.01286v4

cs.CL, cs.AI, cs.CV, cs.HC, cs.LG

CharacterEval: A Chinese Benchmark for Role-Playing Conversational Agent Evaluation

Quan Tu, Shilong Fan, Zihang Tian, Rui Yan

Recently, the advent of large language models (LLMs) has revolutionized generative agents. Among them, Role-Playing Conversational Agents (RPCAs) attract considerable attention due to their ability to emotionally engage users. However, the absence of a comprehensive benchmark impedes progress in this field. To bridge this gap, we introduce CharacterEval, a Chinese benchmark for comprehensive RPCA assessment, complemented by a tailored high-quality dataset. The dataset comprises 1,785 multi-turn role-playing dialogues, encompassing 23,020 examples and featuring 77 characters derived from Chinese novels and scripts. It was carefully constructed, beginning with initial dialogue extraction via GPT-4, followed by rigorous human-led quality control, and enhanced with in-depth character profiles sourced from Baidu Baike. CharacterEval employs a multifaceted evaluation approach, encompassing thirteen targeted metrics on four dimensions. Comprehensive experiments on CharacterEval demonstrate that Chinese LLMs exhibit more promising capabilities than GPT-4 in Chinese role-playing conversation. Source code, data source and reward model will be publicly accessible at https://github.com/morecry/CharacterEval.

http://arxiv.org/abs/2401.01275v2

cs.CL

Zero-Shot Position Debiasing for Large Language Models

Zhongkun Liu, Zheng Chen, Mengqi Zhang, Zhaochun Ren, Pengjie Ren, Zhumin Chen

Fine-tuning has been demonstrated to be an effective method to improve the domain performance of large language models (LLMs). However, LLMs might fit the dataset bias and shortcuts for prediction, leading to poor generation performance. Previous works have proven that LLMs are prone to exhibit position bias, i.e., leveraging information positioned at the beginning or end, or specific positional cues within the input. Existing debiasing methods for LLMs require external bias knowledge or annotated non-biased samples, which is lacking for position debiasing and impractical in reality. In this work, we propose a zero-shot position debiasing (ZOE) framework to mitigate position bias for LLMs. ZOE leverages unsupervised responses from pre-trained LLMs for debiasing without relying on any external knowledge. To improve the quality of unsupervised responses, we propose a MSA module to prune these responses. Experiments on eight datasets and five tasks show that ZOE consistently outperforms existing methods in mitigating three types of position biases. Besides, ZOE achieves this by sacrificing only a small performance on biased samples, which is general and effective. To facilitate the reproducibility of the results, we share the code of all methods and datasets on https://anonymous.4open.science/r/ZOE-F06B.

http://arxiv.org/abs/2401.01218v2

cs.CL, cs.AI, cs.LG, I.2.7

Uncertainty Resolution in Misinformation Detection

Yury Orlovskiy, Camille Thibault, Anne Imouza, Jean-François Godbout, Reihaneh Rabbany, Kellin Pelrine

Misinformation poses a variety of risks, such as undermining public trust and distorting factual discourse. Large Language Models (LLMs) like GPT-4 have been shown effective in mitigating misinformation, particularly in handling statements where enough context is provided. However, they struggle to assess ambiguous or context-deficient statements accurately. This work introduces a new method to resolve uncertainty in such statements. We propose a framework to categorize missing information and publish category labels for the LIAR-New dataset, which is adaptable to cross-domain content with missing information. We then leverage this framework to generate effective user queries for missing context. Compared to baselines, our method improves the rate at which generated questions are answerable by the user by 38 percentage points and classification performance by over 10 percentage points macro F1. Thus, this approach may provide a valuable component for future misinformation mitigation pipelines.

http://arxiv.org/abs/2401.01197v1

cs.CL, cs.AI

SSP: A Simple and Safe automatic Prompt engineering method towards realistic image synthesis on LVM

Weijin Cheng, Jianzhi Liu, Jiawen Deng, Fuji Ren

Recently, text-to-image (T2I) synthesis has undergone significant advancements, particularly with the emergence of Large Language Models (LLM) and their enhancement in Large Vision Models (LVM), greatly enhancing the instruction-following capabilities of traditional T2I models. Nevertheless, previous methods focus on improving generation quality but introduce unsafe factors into prompts. We explore that appending specific camera descriptions to prompts can enhance safety performance. Consequently, we propose a simple and safe prompt engineering method (SSP) to improve image generation quality by providing optimal camera descriptions. Specifically, we create a dataset from multi-datasets as original prompts. To select the optimal camera, we design an optimal camera matching approach and implement a classifier for original prompts capable of automatically matching. Appending camera descriptions to original prompts generates optimized prompts for further LVM image generation. Experiments demonstrate that SSP improves semantic consistency by an average of 16% compared to others and safety metrics by 48.9%.

http://arxiv.org/abs/2401.01128v1

cs.CV

BEV-CLIP: Multi-modal BEV Retrieval Methodology for Complex Scene in Autonomous Driving

Dafeng Wei, Tian Gao, Zhengyu Jia, Changwei Cai, Chengkai Hou, Peng Jia, Fu Liu, Kun Zhan, Jingchen Fan, Yixing Zhao, Yang Wang

The demand for the retrieval of complex scene data in autonomous driving is increasing, especially as passenger vehicles have been equipped with the ability to navigate urban settings, with the imperative to address long-tail scenarios. Meanwhile, under the pre-existing two dimensional image retrieval method, some problems may arise with scene retrieval, such as lack of global feature representation and subpar text retrieval ability. To address these issues, we have proposed \textbf{BEV-CLIP}, the first multimodal Bird's-Eye View(BEV) retrieval methodology that utilizes descriptive text as an input to retrieve corresponding scenes. This methodology applies the semantic feature extraction abilities of a large language model (LLM) to facilitate zero-shot retrieval of extensive text descriptions, and incorporates semi-structured information from a knowledge graph to improve the semantic richness and variety of the language embedding. Our experiments result in 87.66% accuracy on NuScenes dataset in text-to-BEV feature retrieval. The demonstrated cases in our paper support that our retrieval method is also indicated to be effective in identifying certain long-tail corner scenes.

http://arxiv.org/abs/2401.01065v1

cs.CV, cs.AI, I.2.10

Large Language Models in Mental Health Care: a Scoping Review

Yining Hua, Fenglin Liu, Kailai Yang, Zehan Li, Yi-han Sheu, Peilin Zhou, Lauren V. Moran, Sophia Ananiadou, Andrew Beam

Objective: The growing use of large language models (LLMs) stimulates a need for a comprehensive review of their applications and outcomes in mental health care contexts. This scoping review aims to critically analyze the existing development and applications of LLMs in mental health care, highlighting their successes and identifying their challenges and limitations in these specialized fields. Materials and Methods: A broad literature search was conducted in November 2023 using six databases (PubMed, Web of Science, Google Scholar, arXiv, medRxiv, and PsyArXiv) following the 2020 version of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A total of 313 publications were initially identified, and after applying the study inclusion criteria, 34 publications were selected for the final review. Results: We identified diverse applications of LLMs in mental health care, including diagnosis, therapy, patient engagement enhancement, etc. Key challenges include data availability and reliability, nuanced handling of mental states, and effective evaluation methods. Despite successes in accuracy and accessibility improvement, gaps in clinical applicability and ethical considerations were evident, pointing to the need for robust data, standardized evaluations, and interdisciplinary collaboration. Conclusion: LLMs show promising potential in advancing mental health care, with applications in diagnostics, and patient support. Continued advancements depend on collaborative, multidisciplinary efforts focused on framework enhancement, rigorous dataset development, technological refinement, and ethical integration to ensure the effective and safe application of LLMs in mental health care.

http://arxiv.org/abs/2401.02984v1

cs.CL, cs.AI

Astraios: Parameter-Efficient Instruction Tuning Code Large Language Models

Terry Yue Zhuo, Armel Zebaze, Nitchakarn Suppattarachai, Leandro von Werra, Harm de Vries, Qian Liu, Niklas Muennighoff

The high cost of full-parameter fine-tuning (FFT) of Large Language Models (LLMs) has led to a series of parameter-efficient fine-tuning (PEFT) methods. However, it remains unclear which methods provide the best cost-performance trade-off at different model scales. We introduce Astraios, a suite of 28 instruction-tuned OctoCoder models using 7 tuning methods and 4 model sizes up to 16 billion parameters. Through investigations across 5 tasks and 8 different datasets encompassing both code comprehension and code generation tasks, we find that FFT generally leads to the best downstream performance across all scales, and PEFT methods differ significantly in their efficacy based on the model scale. LoRA usually offers the most favorable trade-off between cost and performance. Further investigation into the effects of these methods on both model robustness and code security reveals that larger models tend to demonstrate reduced robustness and less security. At last, we explore the relationships among updated parameters, cross-entropy loss, and task performance. We find that the tuning effectiveness observed in small models generalizes well to larger models, and the validation loss in instruction tuning can be a reliable indicator of overall downstream performance.

http://arxiv.org/abs/2401.00788v1

cs.CL, cs.AI, cs.SE

BIBench: Benchmarking Data Analysis Knowledge of Large Language Models

Shu Liu, Shangqing Zhao, Chenghao Jia, Xinlin Zhuang, Zhaoguang Long, Qingquan Wu, Chong Yang, Aimin Zhou, Man Lan

Large Language Models (LLMs) have demonstrated impressive capabilities across a wide range of tasks. However, their proficiency and reliability in the specialized domain of Data Analysis, particularly with a focus on data-driven thinking, remain uncertain. To bridge this gap, we introduce BIBench, a comprehensive benchmark designed to evaluate the data analysis capabilities of LLMs within the context of Business Intelligence (BI). BIBench assesses LLMs across three dimensions: 1) BI foundational knowledge, evaluating the models' numerical reasoning and familiarity with financial concepts; 2) BI knowledge application, determining the models' ability to quickly comprehend textual information and generate analysis questions from multiple views; and 3) BI technical skills, examining the models' use of technical knowledge to address real-world data analysis challenges. BIBench comprises 11 sub-tasks, spanning three categories of task types: classification, extraction, and generation. Additionally, we've developed BIChat, a domain-specific dataset with over a million data points, to fine-tune LLMs. We will release BIBenchmark, BIChat, and the evaluation scripts at \url{https://github.com/cubenlp/BIBench}. This benchmark aims to provide a measure for in-depth analysis of LLM abilities and foster the advancement of LLMs in the field of data analysis.

http://arxiv.org/abs/2401.02982v3

cs.CL, cs.AI

Fine-tuning and Utilization Methods of Domain-specific LLMs

Cheonsu Jeong

Recent releases of pre-trained Large Language Models (LLMs) have gained considerable traction, yet research on fine-tuning and employing domain-specific LLMs remains scarce. This study investigates approaches for fine-tuning and leveraging domain-specific LLMs, highlighting trends in LLMs, foundational models, and methods for domain-specific pre-training. Focusing on the financial sector, it details dataset selection, preprocessing, model choice, and considerations crucial for LLM fine-tuning in finance. Addressing the unique characteristics of financial data, the study explores the construction of domain-specific vocabularies and considerations for security and regulatory compliance. In the practical application of LLM fine-tuning, the study outlines the procedure and implementation for generating domain-specific LLMs in finance. Various financial cases, including stock price prediction, sentiment analysis of financial news, automated document processing, research, information extraction, and customer service enhancement, are exemplified. The study explores the potential of LLMs in the financial domain, identifies limitations, and proposes directions for improvement, contributing valuable insights for future research. Ultimately, it advances natural language processing technology in business, suggesting proactive LLM utilization in financial services across industries.

http://arxiv.org/abs/2401.02981v2

cs.CL, cs.AI

Digger: Detecting Copyright Content Mis-usage in Large Language Model Training

Haodong Li, Gelei Deng, Yi Liu, Kailong Wang, Yuekang Li, Tianwei Zhang, Yang Liu, Guoai Xu, Guosheng Xu, Haoyu Wang

Pre-training, which utilizes extensive and varied datasets, is a critical factor in the success of Large Language Models (LLMs) across numerous applications. However, the detailed makeup of these datasets is often not disclosed, leading to concerns about data security and potential misuse. This is particularly relevant when copyrighted material, still under legal protection, is used inappropriately, either intentionally or unintentionally, infringing on the rights of the authors. In this paper, we introduce a detailed framework designed to detect and assess the presence of content from potentially copyrighted books within the training datasets of LLMs. This framework also provides a confidence estimation for the likelihood of each content sample's inclusion. To validate our approach, we conduct a series of simulated experiments, the results of which affirm the framework's effectiveness in identifying and addressing instances of content misuse in LLM training processes. Furthermore, we investigate the presence of recognizable quotes from famous literary works within these datasets. The outcomes of our study have significant implications for ensuring the ethical use of copyrighted materials in the development of LLMs, highlighting the need for more transparent and responsible data management practices in this field.

http://arxiv.org/abs/2401.00676v1

cs.CR, cs.CL, cs.LG

Beyond Efficiency: A Systematic Survey of Resource-Efficient Large Language Models

Guangji Bai, Zheng Chai, Chen Ling, Shiyu Wang, Jiaying Lu, Nan Zhang, Tingwei Shi, Ziyang Yu, Mengdan Zhu, Yifei Zhang, Carl Yang, Yue Cheng, Liang Zhao

The burgeoning field of Large Language Models (LLMs), exemplified by sophisticated models like OpenAI's ChatGPT, represents a significant advancement in artificial intelligence. These models, however, bring forth substantial challenges in the high consumption of computational, memory, energy, and financial resources, especially in environments with limited resource capabilities. This survey aims to systematically address these challenges by reviewing a broad spectrum of techniques designed to enhance the resource efficiency of LLMs. We categorize methods based on their optimization focus: computational, memory, energy, financial, and network resources and their applicability across various stages of an LLM's lifecycle, including architecture design, pretraining, finetuning, and system design. Additionally, the survey introduces a nuanced categorization of resource efficiency techniques by their specific resource types, which uncovers the intricate relationships and mappings between various resources and corresponding optimization techniques. A standardized set of evaluation metrics and datasets is also presented to facilitate consistent and fair comparisons across different models and techniques. By offering a comprehensive overview of the current sota and identifying open research avenues, this survey serves as a foundational reference for researchers and practitioners, aiding them in developing more sustainable and efficient LLMs in a rapidly evolving landscape.

http://arxiv.org/abs/2401.00625v2

cs.LG

DocLLM: A layout-aware generative language model for multimodal document understanding

Dongsheng Wang, Natraj Raman, Mathieu Sibue, Zhiqiang Ma, Petr Babkin, Simerjot Kaur, Yulong Pei, Armineh Nourbakhsh, Xiaomo Liu

Enterprise documents such as forms, invoices, receipts, reports, contracts, and other similar records, often carry rich semantics at the intersection of textual and spatial modalities. The visual cues offered by their complex layouts play a crucial role in comprehending these documents effectively. In this paper, we present DocLLM, a lightweight extension to traditional large language models (LLMs) for reasoning over visual documents, taking into account both textual semantics and spatial layout. Our model differs from existing multimodal LLMs by avoiding expensive image encoders and focuses exclusively on bounding box information to incorporate the spatial layout structure. Specifically, the cross-alignment between text and spatial modalities is captured by decomposing the attention mechanism in classical transformers to a set of disentangled matrices. Furthermore, we devise a pre-training objective that learns to infill text segments. This approach allows us to address irregular layouts and heterogeneous content frequently encountered in visual documents. The pre-trained model is fine-tuned using a large-scale instruction dataset, covering four core document intelligence tasks. We demonstrate that our solution outperforms SotA LLMs on 14 out of 16 datasets across all tasks, and generalizes well to 4 out of 5 previously unseen datasets.

http://arxiv.org/abs/2401.00908v1

cs.CL

LaFFi: Leveraging Hybrid Natural Language Feedback for Fine-tuning Language Models

Qianxi Li, Yingyue Cao, Jikun Kang, Tianpei Yang, Xi Chen, Jun Jin, Matthew E. Taylor

Fine-tuning Large Language Models (LLMs) adapts a trained model to specific downstream tasks, significantly improving task-specific performance. Supervised Fine-Tuning (SFT) is a common approach, where an LLM is trained to produce desired answers. However, LLMs trained with SFT sometimes make simple mistakes and result in hallucinations on reasoning tasks such as question-answering. Without external feedback, it is difficult for SFT to learn a good mapping between the question and the desired answer, especially with a small dataset. This paper introduces an alternative to SFT called Natural Language Feedback for Finetuning LLMs (LaFFi). LaFFi has LLMs directly predict the feedback they will receive from an annotator. We find that requiring such reflection can significantly improve the accuracy in in-domain question-answering tasks, providing a promising direction for the application of natural language feedback in the realm of SFT LLMs. Additional ablation studies show that the portion of human-annotated data in the annotated datasets affects the fine-tuning performance.

http://arxiv.org/abs/2401.00907v1

cs.LG, cs.AI, cs.CL

Exploring the Effectiveness of Instruction Tuning in Biomedical Language Processing

Omid Rohanian, Mohammadmahdi Nouriborji, David A. Clifton

Large Language Models (LLMs), particularly those similar to ChatGPT, have significantly influenced the field of Natural Language Processing (NLP). While these models excel in general language tasks, their performance in domain-specific downstream tasks such as biomedical and clinical Named Entity Recognition (NER), Relation Extraction (RE), and Medical Natural Language Inference (NLI) is still evolving. In this context, our study investigates the potential of instruction tuning for biomedical language processing, applying this technique to two general LLMs of substantial scale. We present a comprehensive, instruction-based model trained on a dataset that consists of approximately $200,000$ instruction-focused samples. This dataset represents a carefully curated compilation of existing data, meticulously adapted and reformatted to align with the specific requirements of our instruction-based tasks. This initiative represents an important step in utilising such models to achieve results on par with specialised encoder-only models like BioBERT and BioClinicalBERT for various classical biomedical NLP tasks. Our work includes an analysis of the dataset's composition and its impact on model performance, providing insights into the intricacies of instruction tuning. By sharing our codes, models, and the distinctively assembled instruction-based dataset, we seek to encourage ongoing research and development in this area.

http://arxiv.org/abs/2401.00579v1

cs.CL, cs.AI, cs.LG, 68T50, I.2.7

E-chat: Emotion-sensitive Spoken Dialogue System with Large Language Models

Hongfei Xue, Yuhao Liang, Bingshen Mu, Shiliang Zhang, Mengzhe Chen, Qian Chen, Lei Xie

This study focuses on emotion-sensitive spoken dialogue in human-machine speech interaction. With the advancement of Large Language Models (LLMs), dialogue systems can handle multimodal data, including audio. Recent models have enhanced the understanding of complex audio signals through the integration of various audio events. However, they are unable to generate appropriate responses based on emotional speech. To address this, we introduce the Emotional chat Model (E-chat), a novel spoken dialogue system capable of comprehending and responding to emotions conveyed from speech. This model leverages an emotion embedding extracted by a speech encoder, combined with LLMs, enabling it to respond according to different emotional contexts. Additionally, we introduce the E-chat200 dataset, designed explicitly for emotion-sensitive spoken dialogue. In various evaluation metrics, E-chat consistently outperforms baseline LLMs, demonstrating its potential in emotional comprehension and human-machine interaction.

http://arxiv.org/abs/2401.00475v2

cs.SD, eess.AS

GeoGalactica: A Scientific Large Language Model in Geoscience

Zhouhan Lin, Cheng Deng, Le Zhou, Tianhang Zhang, Yi Xu, Yutong Xu, Zhongmou He, Yuanyuan Shi, Beiya Dai, Yunchong Song, Boyi Zeng, Qiyuan Chen, Yuxun Miao, Bo Xue, Shu Wang, Luoyi Fu, Weinan Zhang, Junxian He, Yunqiang Zhu, Xinbing Wang, Chenghu Zhou

Large language models (LLMs) have achieved huge success for their general knowledge and ability to solve a wide spectrum of tasks in natural language processing (NLP). Due to their impressive abilities, LLMs have shed light on potential inter-discipline applications to foster scientific discoveries of a specific domain by using artificial intelligence (AI for science, AI4S). In the meantime, utilizing NLP techniques in geoscience research and practice is wide and convoluted, contributing from knowledge extraction and document classification to question answering and knowledge discovery. In this work, we take the initial step to leverage LLM for science, through a rather straightforward approach. We try to specialize an LLM into geoscience, by further pre-training the model with a vast amount of texts in geoscience, as well as supervised fine-tuning (SFT) the resulting model with our custom collected instruction tuning dataset. These efforts result in a model GeoGalactica consisting of 30 billion parameters. To our best knowledge, it is the largest language model for the geoscience domain. More specifically, GeoGalactica is from further pre-training of Galactica. We train GeoGalactica over a geoscience-related text corpus containing 65 billion tokens, preserving as the largest geoscience-specific text corpus. Then we fine-tune the model with 1 million pairs of instruction-tuning data consisting of questions that demand professional geoscience knowledge to answer. In this technical report, we will illustrate in detail all aspects of GeoGalactica, including data collection, data cleaning, base model selection, pre-training, SFT, and evaluation. We open-source our data curation tools and the checkpoints of GeoGalactica during the first 3/4 of pre-training.

http://arxiv.org/abs/2401.00434v2

cs.CL, I.2.7; F.4.1

keqing: knowledge-based question answering is a nature chain-of-thought mentor of LLM

Chaojie Wang, Yishi Xu, Zhong Peng, Chenxi Zhang, Bo Chen, Xinrun Wang, Lei Feng, Bo An

Large language models (LLMs) have exhibited remarkable performance on various natural language processing (NLP) tasks, especially for question answering. However, in the face of problems beyond the scope of knowledge, these LLMs tend to talk nonsense with a straight face, where the potential solution could be incorporating an Information Retrieval (IR) module and generating response based on these retrieved knowledge. In this paper, we present a novel framework to assist LLMs, such as ChatGPT, to retrieve question-related structured information on the knowledge graph, and demonstrate that Knowledge-based question answering (Keqing) could be a nature Chain-of-Thought (CoT) mentor to guide the LLM to sequentially find the answer entities of a complex question through interpretable logical chains. Specifically, the workflow of Keqing will execute decomposing a complex question according to predefined templates, retrieving candidate entities on knowledge graph, reasoning answers of sub-questions, and finally generating response with reasoning paths, which greatly improves the reliability of LLM's response. The experimental results on KBQA datasets show that Keqing can achieve competitive performance and illustrate the logic of answering each question.

http://arxiv.org/abs/2401.00426v1

cs.CL, cs.AI

RAGTruth: A Hallucination Corpus for Developing Trustworthy Retrieval-Augmented Language Models

Yuanhao Wu, Juno Zhu, Siliang Xu, Kashun Shum, Cheng Niu, Randy Zhong, Juntong Song, Tong Zhang

Retrieval-augmented generation (RAG) has become a main technique for alleviating hallucinations in large language models (LLMs). Despite the integration of RAG, LLMs may still present unsupported or contradictory claims to the retrieved contents. In order to develop effective hallucination prevention strategies under RAG, it is important to create benchmark datasets that can measure the extent of hallucination. This paper presents RAGTruth, a corpus tailored for analyzing word-level hallucinations in various domains and tasks within the standard RAG frameworks for LLM applications. RAGTruth comprises nearly 18,000 naturally generated responses from diverse LLMs using RAG. These responses have undergone meticulous manual annotations at both the individual cases and word levels, incorporating evaluations of hallucination intensity. We not only benchmark hallucination frequencies across different LLMs, but also critically assess the effectiveness of several existing hallucination detection methodologies. Furthermore, we show that using a high-quality dataset such as RAGTruth, it is possible to finetune a relatively small LLM and achieve a competitive level of performance in hallucination detection when compared to the existing prompt-based approaches using state-of-the-art large language models such as GPT-4.

http://arxiv.org/abs/2401.00396v1

cs.CL

Improving Text Embeddings with Large Language Models

Liang Wang, Nan Yang, Xiaolong Huang, Linjun Yang, Rangan Majumder, Furu Wei

In this paper, we introduce a novel and simple method for obtaining high-quality text embeddings using only synthetic data and less than 1k training steps. Unlike existing methods that often depend on multi-stage intermediate pre-training with billions of weakly-supervised text pairs, followed by fine-tuning with a few labeled datasets, our method does not require building complex training pipelines or relying on manually collected datasets that are often constrained by task diversity and language coverage. We leverage proprietary LLMs to generate diverse synthetic data for hundreds of thousands of text embedding tasks across nearly 100 languages. We then fine-tune open-source decoder-only LLMs on the synthetic data using standard contrastive loss. Experiments demonstrate that our method achieves strong performance on highly competitive text embedding benchmarks without using any labeled data. Furthermore, when fine-tuned with a mixture of synthetic and labeled data, our model sets new state-of-the-art results on the BEIR and MTEB benchmarks.

http://arxiv.org/abs/2401.00368v2

cs.CL, cs.IR

Uncertainty-Penalized Reinforcement Learning from Human Feedback with Diverse Reward LoRA Ensembles

Yuanzhao Zhai, Han Zhang, Yu Lei, Yue Yu, Kele Xu, Dawei Feng, Bo Ding, Huaimin Wang

Reinforcement learning from human feedback (RLHF) emerges as a promising paradigm for aligning large language models (LLMs). However, a notable challenge in RLHF is overoptimization, where beyond a certain threshold, the pursuit of higher rewards leads to a decline in human preferences. In this paper, we observe the weakness of KL regularization which is commonly employed in existing RLHF methods to address overoptimization. To mitigate this limitation, we scrutinize the RLHF objective in the offline dataset and propose uncertainty-penalized RLHF (UP-RLHF), which incorporates uncertainty regularization during RL-finetuning. To enhance the uncertainty quantification abilities for reward models, we first propose a diverse low-rank adaptation (LoRA) ensemble by maximizing the nuclear norm of LoRA matrix concatenations. Then we optimize policy models utilizing penalized rewards, determined by both rewards and uncertainties provided by the diverse reward LoRA ensembles. Our experimental results, based on two real human preference datasets, showcase the effectiveness of diverse reward LoRA ensembles in quantifying reward uncertainty. Additionally, uncertainty regularization in UP-RLHF proves to be pivotal in mitigating overoptimization, thereby contributing to the overall performance.

http://arxiv.org/abs/2401.00243v1

cs.LG

KAXAI: An Integrated Environment for Knowledge Analysis and Explainable AI

Saikat Barua, Dr. Sifat Momen

In order to fully harness the potential of machine learning, it is crucial to establish a system that renders the field more accessible and less daunting for individuals who may not possess a comprehensive understanding of its intricacies. The paper describes the design of a system that integrates AutoML, XAI, and synthetic data generation to provide a great UX design for users. The system allows users to navigate and harness the power of machine learning while abstracting its complexities and providing high usability. The paper proposes two novel classifiers, Logistic Regression Forest and Support Vector Tree, for enhanced model performance, achieving 96\% accuracy on a diabetes dataset and 93\% on a survey dataset. The paper also introduces a model-dependent local interpreter called MEDLEY and evaluates its interpretation against LIME, Greedy, and Parzen. Additionally, the paper introduces LLM-based synthetic data generation, library-based data generation, and enhancing the original dataset with GAN. The findings on synthetic data suggest that enhancing the original dataset with GAN is the most reliable way to generate synthetic data, as evidenced by KS tests, standard deviation, and feature importance. The authors also found that GAN works best for quantitative datasets.

http://arxiv.org/abs/2401.00193v1

cs.LG, cs.HC

Pushing Boundaries: Exploring Zero Shot Object Classification with Large Multimodal Models

Ashhadul Islam, Md. Rafiul Biswas, Wajdi Zaghouani, Samir Brahim Belhaouari, Zubair Shah

$ $The synergy of language and vision models has given rise to Large Language and Vision Assistant models (LLVAs), designed to engage users in rich conversational experiences intertwined with image-based queries. These comprehensive multimodal models seamlessly integrate vision encoders with Large Language Models (LLMs), expanding their applications in general-purpose language and visual comprehension. The advent of Large Multimodal Models (LMMs) heralds a new era in Artificial Intelligence (AI) assistance, extending the horizons of AI utilization. This paper takes a unique perspective on LMMs, exploring their efficacy in performing image classification tasks using tailored prompts designed for specific datasets. We also investigate the LLVAs zero-shot learning capabilities. Our study includes a benchmarking analysis across four diverse datasets: MNIST, Cats Vs. Dogs, Hymnoptera (Ants Vs. Bees), and an unconventional dataset comprising Pox Vs. Non-Pox skin images. The results of our experiments demonstrate the model's remarkable performance, achieving classification accuracies of 85\%, 100\%, 77\%, and 79\% for the respective datasets without any fine-tuning. To bolster our analysis, we assess the model's performance post fine-tuning for specific tasks. In one instance, fine-tuning is conducted over a dataset comprising images of faces of children with and without autism. Prior to fine-tuning, the model demonstrated a test accuracy of 55\%, which significantly improved to 83\% post fine-tuning. These results, coupled with our prior findings, underscore the transformative potential of LLVAs and their versatile applications in real-world scenarios.

http://arxiv.org/abs/2401.00127v1

cs.CV, cs.SI

Teach Large Language Models to Forget Privacy

Ran Yan, Yujun Li, Wenqian Li, Peihua Mai, Yan Pang, Yinchuan Li

Large Language Models (LLMs) have proven powerful, but the risk of privacy leakage remains a significant concern. Traditional privacy-preserving methods, such as Differential Privacy and Homomorphic Encryption, are inadequate for black-box API-only settings, demanding either model transparency or heavy computational resources. We propose Prompt2Forget (P2F), the first framework designed to tackle the LLM local privacy challenge by teaching LLM to forget. The method involves decomposing full questions into smaller segments, generating fabricated answers, and obfuscating the model's memory of the original input. A benchmark dataset was crafted with questions containing privacy-sensitive information from diverse fields. P2F achieves zero-shot generalization, allowing adaptability across a wide range of use cases without manual adjustments. Experimental results indicate P2F's robust capability to obfuscate LLM's memory, attaining a forgetfulness score of around 90\% without any utility loss. This represents an enhancement of up to 63\% when contrasted with the naive direct instruction technique, highlighting P2F's efficacy in mitigating memory retention of sensitive information within LLMs. Our findings establish the first benchmark in the novel field of the LLM forgetting task, representing a meaningful advancement in privacy preservation in the emerging LLM domain.

http://arxiv.org/abs/2401.00870v1

cs.CR, cs.AI, I.2.7

K-PERM: Personalized Response Generation Using Dynamic Knowledge Retrieval and Persona-Adaptive Queries

Kanak Raj, Kaushik Roy, Vamshi Bonagiri, Priyanshul Govil, Krishnaprasad Thirunarayanan, Manas Gaur

Personalizing conversational agents can enhance the quality of conversations and increase user engagement. However, they often lack external knowledge to appropriately tend to a user's persona. This is particularly crucial for practical applications like mental health support, nutrition planning, culturally sensitive conversations, or reducing toxic behavior in conversational agents. To enhance the relevance and comprehensiveness of personalized responses, we propose using a two-step approach that involves (1) selectively integrating user personas and (2) contextualizing the response with supplementing information from a background knowledge source. We develop K-PERM (Knowledge-guided PErsonalization with Reward Modulation), a dynamic conversational agent that combines these elements. K-PERM achieves state-of-the-art performance on the popular FoCus dataset, containing real-world personalized conversations concerning global landmarks. We show that using responses from K-PERM can improve performance in state-of-the-art LLMs (GPT 3.5) by 10.5%, highlighting the impact of K-PERM for personalizing chatbots.

http://arxiv.org/abs/2312.17748v2

cs.IR

Jatmo: Prompt Injection Defense by Task-Specific Finetuning

Julien Piet, Maha Alrashed, Chawin Sitawarin, Sizhe Chen, Zeming Wei, Elizabeth Sun, Basel Alomair, David Wagner

Large Language Models (LLMs) are attracting significant research attention due to their instruction-following abilities, allowing users and developers to leverage LLMs for a variety of tasks. However, LLMs are vulnerable to prompt-injection attacks: a class of attacks that hijack the model's instruction-following abilities, changing responses to prompts to undesired, possibly malicious ones. In this work, we introduce Jatmo, a method for generating task-specific models resilient to prompt-injection attacks. Jatmo leverages the fact that LLMs can only follow instructions once they have undergone instruction tuning. It harnesses a teacher instruction-tuned model to generate a task-specific dataset, which is then used to fine-tune a base model (i.e., a non-instruction-tuned model). Jatmo only needs a task prompt and a dataset of inputs for the task: it uses the teacher model to generate outputs. For situations with no pre-existing datasets, Jatmo can use a single example, or in some cases none at all, to produce a fully synthetic dataset. Our experiments on seven tasks show that Jatmo models provide similar quality of outputs on their specific task as standard LLMs, while being resilient to prompt injections. The best attacks succeeded in less than 0.5% of cases against our models, versus 87% success rate against GPT-3.5-Turbo. We release Jatmo at https://github.com/wagner-group/prompt-injection-defense.

http://arxiv.org/abs/2312.17673v2

cs.CR, cs.AI, cs.CL

Gemini in Reasoning: Unveiling Commonsense in Multimodal Large Language Models

Yuqing Wang, Yun Zhao

The burgeoning interest in Multimodal Large Language Models (MLLMs), such as OpenAI's GPT-4V(ision), has significantly impacted both academic and industrial realms. These models enhance Large Language Models (LLMs) with advanced visual understanding capabilities, facilitating their application in a variety of multimodal tasks. Recently, Google introduced Gemini, a cutting-edge MLLM designed specifically for multimodal integration. Despite its advancements, preliminary benchmarks indicate that Gemini lags behind GPT models in commonsense reasoning tasks. However, this assessment, based on a limited dataset (i.e., HellaSWAG), does not fully capture Gemini's authentic commonsense reasoning potential. To address this gap, our study undertakes a thorough evaluation of Gemini's performance in complex reasoning tasks that necessitate the integration of commonsense knowledge across modalities. We carry out a comprehensive analysis of 12 commonsense reasoning datasets, ranging from general to domain-specific tasks. This includes 11 datasets focused solely on language, as well as one that incorporates multimodal elements. Our experiments across four LLMs and two MLLMs demonstrate Gemini's competitive commonsense reasoning capabilities. Additionally, we identify common challenges faced by current LLMs and MLLMs in addressing commonsense problems, underscoring the need for further advancements in enhancing the commonsense reasoning abilities of these models.

http://arxiv.org/abs/2312.17661v1

cs.CL, cs.AI, cs.CV

Building Efficient Universal Classifiers with Natural Language Inference

Moritz Laurer, Wouter van Atteveldt, Andreu Casas, Kasper Welbers

Generative Large Language Models (LLMs) have become the mainstream choice for fewshot and zeroshot learning thanks to the universality of text generation. Many users, however, do not need the broad capabilities of generative LLMs when they only want to automate a classification task. Smaller BERT-like models can also learn universal tasks, which allow them to do any text classification task without requiring fine-tuning (zeroshot classification) or to learn new tasks with only a few examples (fewshot), while being significantly more efficient than generative LLMs. This paper (1) explains how Natural Language Inference (NLI) can be used as a universal classification task that follows similar principles as instruction fine-tuning of generative LLMs, (2) provides a step-by-step guide with reusable Jupyter notebooks for building a universal classifier, and (3) shares the resulting universal classifier that is trained on 33 datasets with 389 diverse classes. Parts of the code we share has been used to train our older zeroshot classifiers that have been downloaded more than 55 million times via the Hugging Face Hub as of December 2023. Our new classifier improves zeroshot performance by 9.4%.

http://arxiv.org/abs/2312.17543v2

cs.CL, cs.AI

Overview of the PromptCBLUE Shared Task in CHIP2023

Wei Zhu, Xiaoling Wang, Mosha Chen, Buzhou Tang

This paper presents an overview of the PromptCBLUE shared task (http://cips-chip.org.cn/2023/eval1) held in the CHIP-2023 Conference. This shared task reformualtes the CBLUE benchmark, and provide a good testbed for Chinese open-domain or medical-domain large language models (LLMs) in general medical natural language processing. Two different tracks are held: (a) prompt tuning track, investigating the multitask prompt tuning of LLMs, (b) probing the in-context learning capabilities of open-sourced LLMs. Many teams from both the industry and academia participated in the shared tasks, and the top teams achieved amazing test results. This paper describes the tasks, the datasets, evaluation metrics, and the top systems for both tasks. Finally, the paper summarizes the techniques and results of the evaluation of the various approaches explored by the participating teams.

http://arxiv.org/abs/2312.17522v1

cs.CL

Differentially Private Low-Rank Adaptation of Large Language Model Using Federated Learning

Xiao-Yang Liu, Rongyi Zhu, Daochen Zha, Jiechao Gao, Shan Zhong, Meikang Qiu

The surge in interest and application of large language models (LLMs) has sparked a drive to fine-tune these models to suit specific applications, such as finance and medical science. However, concerns regarding data privacy have emerged, especially when multiple stakeholders aim to collaboratively enhance LLMs using sensitive data. In this scenario, federated learning becomes a natural choice, allowing decentralized fine-tuning without exposing raw data to central servers. Motivated by this, we investigate how data privacy can be ensured in LLM fine-tuning through practical federated learning approaches, enabling secure contributions from multiple parties to enhance LLMs. Yet, challenges arise: 1) despite avoiding raw data exposure, there is a risk of inferring sensitive information from model outputs, and 2) federated learning for LLMs incurs notable communication overhead. To address these challenges, this article introduces DP-LoRA, a novel federated learning algorithm tailored for LLMs. DP-LoRA preserves data privacy by employing a Gaussian mechanism that adds noise in weight updates, maintaining individual data privacy while facilitating collaborative model training. Moreover, DP-LoRA optimizes communication efficiency via low-rank adaptation, minimizing the transmission of updated weights during distributed training. The experimental results across medical, financial, and general datasets using various LLMs demonstrate that DP-LoRA effectively ensures strict privacy constraints while minimizing communication overhead.

http://arxiv.org/abs/2312.17493v1

cs.LG, cs.CR

The Right Prompts for the Job: Repair Code-Review Defects with Large Language Model

Zelin Zhao, Zhaogui Xu, Jialong Zhu, Peng Di, Yuan Yao, Xiaoxing Ma

Automatic program repair (APR) techniques have the potential to reduce manual efforts in uncovering and repairing program defects during the code review (CR) process. However, the limited accuracy and considerable time costs associated with existing APR approaches hinder their adoption in industrial practice. One key factor is the under-utilization of review comments, which provide valuable insights into defects and potential fixes. Recent advancements in Large Language Models (LLMs) have enhanced their ability to comprehend natural and programming languages, enabling them to generate patches based on review comments. This paper conducts a comprehensive investigation into the effective utilization of LLMs for repairing CR defects. In this study, various prompts are designed and compared across mainstream LLMs using two distinct datasets from human reviewers and automated checkers. Experimental results demonstrate a remarkable repair rate of 72.97% with the best prompt, highlighting a substantial improvement in the effectiveness and practicality of automatic repair techniques.

http://arxiv.org/abs/2312.17485v1

cs.SE, 68T01, I.2.5; D.2.0

Truth Forest: Toward Multi-Scale Truthfulness in Large Language Models through Intervention without Tuning

Zhongzhi Chen, Xingwu Sun, Xianfeng Jiao, Fengzong Lian, Zhanhui Kang, Di Wang, Cheng-Zhong Xu

Despite the great success of large language models (LLMs) in various tasks, they suffer from generating hallucinations. We introduce Truth Forest, a method that enhances truthfulness in LLMs by uncovering hidden truth representations using multi-dimensional orthogonal probes. Specifically, it creates multiple orthogonal bases for modeling truth by incorporating orthogonal constraints into the probes. Moreover, we introduce Random Peek, a systematic technique considering an extended range of positions within the sequence, reducing the gap between discerning and generating truth features in LLMs. By employing this approach, we improved the truthfulness of Llama-2-7B from 40.8\% to 74.5\% on TruthfulQA. Likewise, significant improvements are observed in fine-tuned models. We conducted a thorough analysis of truth features using probes. Our visualization results show that orthogonal probes capture complementary truth-related features, forming well-defined clusters that reveal the inherent structure of the dataset.

http://arxiv.org/abs/2312.17484v2

cs.CL, cs.AI

EHR Interaction Between Patients and AI: NoteAid EHR Interaction

Xiaocheng Zhang, Zonghai Yao, Hong Yu

With the rapid advancement of Large Language Models (LLMs) and their outstanding performance in semantic and contextual comprehension, the potential of LLMs in specialized domains warrants exploration. This paper introduces the NoteAid EHR Interaction Pipeline, an innovative approach developed using generative LLMs to assist in patient education, a task stemming from the need to aid patients in understanding Electronic Health Records (EHRs). Building upon the NoteAid work, we designed two novel tasks from the patient's perspective: providing explanations for EHR content that patients may not understand and answering questions posed by patients after reading their EHRs. We extracted datasets containing 10,000 instances from MIMIC Discharge Summaries and 876 instances from the MADE medical notes collection, respectively, executing the two tasks through the NoteAid EHR Interaction Pipeline with these data. Performance data of LLMs on these tasks were collected and constructed as the corresponding NoteAid EHR Interaction Dataset. Through a comprehensive evaluation of the entire dataset using LLM assessment and a rigorous manual evaluation of 64 instances, we showcase the potential of LLMs in patient education. Besides, the results provide valuable data support for future exploration and applications in this domain while also supplying high-quality synthetic datasets for in-house system training.

http://arxiv.org/abs/2312.17475v1

cs.CL, cs.AI

Video Understanding with Large Language Models: A Survey

Yunlong Tang, Jing Bi, Siting Xu, Luchuan Song, Susan Liang, Teng Wang, Daoan Zhang, Jie An, Jingyang Lin, Rongyi Zhu, Ali Vosoughi, Chao Huang, Zeliang Zhang, Feng Zheng, Jianguo Zhang, Ping Luo, Jiebo Luo, Chenliang Xu

With the burgeoning growth of online video platforms and the escalating volume of video content, the demand for proficient video understanding tools has intensified markedly. Given the remarkable capabilities of Large Language Models (LLMs) in language and multimodal tasks, this survey provides a detailed overview of the recent advancements in video understanding harnessing the power of LLMs (Vid-LLMs). The emergent capabilities of Vid-LLMs are surprisingly advanced, particularly their ability for open-ended spatial-temporal reasoning combined with commonsense knowledge, suggesting a promising path for future video understanding. We examine the unique characteristics and capabilities of Vid-LLMs, categorizing the approaches into four main types: LLM-based Video Agents, Vid-LLMs Pretraining, Vid-LLMs Instruction Tuning, and Hybrid Methods. Furthermore, this survey presents a comprehensive study of the tasks, datasets, and evaluation methodologies for Vid-LLMs. Additionally, it explores the expansive applications of Vid-LLMs across various domains, highlighting their remarkable scalability and versatility in real-world video understanding challenges. Finally, it summarizes the limitations of existing Vid-LLMs and outlines directions for future research. For more information, readers are recommended to visit the repository at https://github.com/yunlong10/Awesome-LLMs-for-Video-Understanding.

http://arxiv.org/abs/2312.17432v2

cs.CV, cs.CL

AQUALLM: Audio Question Answering Data Generation Using Large Language Models

Swarup Ranjan Behera, Krishna Mohan Injeti, Jaya Sai Kiran Patibandla, Praveen Kumar Pokala, Balakrishna Reddy Pailla

Audio Question Answering (AQA) constitutes a pivotal task in which machines analyze both audio signals and natural language questions to produce precise natural language answers. The significance of possessing high-quality, diverse, and extensive AQA datasets cannot be overstated when aiming for the precision of an AQA system. While there has been notable focus on developing accurate and efficient AQA models, the creation of high-quality, diverse, and extensive datasets for the specific task at hand has not garnered considerable attention. To address this challenge, this work makes several contributions. We introduce a scalable AQA data generation pipeline, denoted as the AQUALLM framework, which relies on Large Language Models (LLMs). This framework utilizes existing audio-caption annotations and incorporates state-of-the-art LLMs to generate expansive, high-quality AQA datasets. Additionally, we present three extensive and high-quality benchmark datasets for AQA, contributing significantly to the progression of AQA research. AQA models trained on the proposed datasets set superior benchmarks compared to the existing state-of-the-art. Moreover, models trained on our datasets demonstrate enhanced generalizability when compared to models trained using human-annotated AQA data. Code and datasets will be accessible on GitHub~\footnote{\url{https://github.com/swarupbehera/AQUALLM}}.

http://arxiv.org/abs/2312.17343v1

cs.CL, cs.AI, cs.LG, cs.MM, cs.SD, eess.AS, I.2.7

Learning Vision from Models Rivals Learning Vision from Data

Yonglong Tian, Lijie Fan, Kaifeng Chen, Dina Katabi, Dilip Krishnan, Phillip Isola

We introduce SynCLR, a novel approach for learning visual representations exclusively from synthetic images and synthetic captions, without any real data. We synthesize a large dataset of image captions using LLMs, then use an off-the-shelf text-to-image model to generate multiple images corresponding to each synthetic caption. We perform visual representation learning on these synthetic images via contrastive learning, treating images sharing the same caption as positive pairs. The resulting representations transfer well to many downstream tasks, competing favorably with other general-purpose visual representation learners such as CLIP and DINO v2 in image classification tasks. Furthermore, in dense prediction tasks such as semantic segmentation, SynCLR outperforms previous self-supervised methods by a significant margin, e.g., improving over MAE and iBOT by 6.2 and 4.3 mIoU on ADE20k for ViT-B/16.

http://arxiv.org/abs/2312.17742v1

cs.CV

A Simple LLM Framework for Long-Range Video Question-Answering

Ce Zhang, Taixi Lu, Md Mohaiminul Islam, Ziyang Wang, Shoubin Yu, Mohit Bansal, Gedas Bertasius

We present LLoVi, a language-based framework for long-range video question-answering (LVQA). Unlike prior long-range video understanding methods, which are often costly and require specialized long-range video modeling design (e.g., memory queues, state-space layers, etc.), our approach uses a frame/clip-level visual captioner (e.g., BLIP2, LaViLa, LLaVA) coupled with a Large Language Model (GPT-3.5, GPT-4) leading to a simple yet surprisingly effective LVQA framework. Specifically, we decompose short and long-range modeling aspects of LVQA into two stages. First, we use a short-term visual captioner to generate textual descriptions of short video clips (0.5-8s in length) densely sampled from a long input video. Afterward, an LLM aggregates the densely extracted short-term captions to perform long-range temporal reasoning needed to understand the whole video and answer a question. To analyze what makes our simple framework so effective, we thoroughly evaluate various components of our system. Our empirical analysis reveals that the choice of the visual captioner and LLM is critical for good LVQA performance. Furthermore, we show that a specialized prompt that asks the LLM first to summarize the noisy short-term visual captions and then answer a given input question leads to a significant LVQA performance boost. On EgoSchema, which is best known as a very long-form video question-answering benchmark, our method achieves 50.3% accuracy, outperforming the previous best-performing approach by 18.1% (absolute gain). In addition, our approach outperforms the previous state-of-the-art by 4.1% and 3.1% on NeXT-QA and IntentQA. We also extend LLoVi to grounded LVQA and show that it outperforms all prior methods on the NeXT-GQA dataset. We will release our code at https://github.com/CeeZh/LLoVi.

http://arxiv.org/abs/2312.17235v2

cs.CV

Non-Vacuous Generalization Bounds for Large Language Models

Sanae Lotfi, Marc Finzi, Yilun Kuang, Tim G. J. Rudner, Micah Goldblum, Andrew Gordon Wilson

Modern language models can contain billions of parameters, raising the question of whether they can generalize beyond the training data or simply regurgitate their training corpora. We provide the first non-vacuous generalization bounds for pretrained large language models (LLMs), indicating that language models are capable of discovering regularities that generalize to unseen data. In particular, we derive a compression bound that is valid for the unbounded log-likelihood loss using prediction smoothing, and we extend the bound to handle subsampling, accelerating bound computation on massive datasets. To achieve the extreme level of compression required for non-vacuous generalization bounds, we devise SubLoRA, a low-dimensional non-linear parameterization. Using this approach, we find that larger models have better generalization bounds and are more compressible than smaller models.

http://arxiv.org/abs/2312.17173v2

stat.ML, cs.LG

Large Language Model for Causal Decision Making

Haitao Jiang, Lin Ge, Yuhe Gao, Jianian Wang, Rui Song

Large Language Models (LLMs) have shown their success in language understanding and reasoning on general topics. However, their capability to perform inference based on user-specified structured data and knowledge in corpus-rare concepts, such as causal decision-making is still limited. In this work, we explore the possibility of fine-tuning an open-sourced LLM into LLM4Causal, which can identify the causal task, execute a corresponding function, and interpret its numerical results based on users' queries and the provided dataset. Meanwhile, we propose a data generation process for more controllable GPT prompting and present two instruction-tuning datasets: (1) Causal-Retrieval-Bench for causal problem identification and input parameter extraction for causal function calling and (2) Causal-Interpret-Bench for in-context causal interpretation. By conducting end-to-end evaluations and two ablation studies, we showed that LLM4Causal can deliver end-to-end solutions for causal problems and provide easy-to-understand answers, which significantly outperforms the baselines.

http://arxiv.org/abs/2312.17122v3

cs.CL, cs.AI, stat.ML

DrugAssist: A Large Language Model for Molecule Optimization

Geyan Ye, Xibao Cai, Houtim Lai, Xing Wang, Junhong Huang, Longyue Wang, Wei Liu, Xiangxiang Zeng

Recently, the impressive performance of large language models (LLMs) on a wide range of tasks has attracted an increasing number of attempts to apply LLMs in drug discovery. However, molecule optimization, a critical task in the drug discovery pipeline, is currently an area that has seen little involvement from LLMs. Most of existing approaches focus solely on capturing the underlying patterns in chemical structures provided by the data, without taking advantage of expert feedback. These non-interactive approaches overlook the fact that the drug discovery process is actually one that requires the integration of expert experience and iterative refinement. To address this gap, we propose DrugAssist, an interactive molecule optimization model which performs optimization through human-machine dialogue by leveraging LLM's strong interactivity and generalizability. DrugAssist has achieved leading results in both single and multiple property optimization, simultaneously showcasing immense potential in transferability and iterative optimization. In addition, we publicly release a large instruction-based dataset called MolOpt-Instructions for fine-tuning language models on molecule optimization tasks. We have made our code and data publicly available at https://github.com/blazerye/DrugAssist, which we hope to pave the way for future research in LLMs' application for drug discovery.

http://arxiv.org/abs/2401.10334v1

q-bio.QM, cs.AI, cs.CL, cs.LG

Large Language Models for Conducting Advanced Text Analytics Information Systems Research

Benjamin M. Ampel, Chi-Heng Yang, James Hu, Hsinchun Chen

The exponential growth of digital content has generated massive textual datasets, necessitating advanced analytical approaches. Large Language Models (LLMs) have emerged as tools capable of processing and extracting insights from massive unstructured textual datasets. However, how to leverage LLMs for text-based Information Systems (IS) research is currently unclear. To assist IS research in understanding how to operationalize LLMs, we propose a Text Analytics for Information Systems Research (TAISR) framework. Our proposed framework provides detailed recommendations grounded in IS and LLM literature on how to conduct meaningful text-based IS research. We conducted three case studies in business intelligence using our TAISR framework to demonstrate its application across several IS research contexts. We also outline potential challenges and limitations in adopting LLMs for IS. By offering a systematic approach and evidence of its utility, our TAISR framework contributes to future IS research streams looking to incorporate powerful LLMs for text analytics.

http://arxiv.org/abs/2312.17278v1

cs.CL, cs.LG

PanGu-$π$: Enhancing Language Model Architectures via Nonlinearity Compensation

Yunhe Wang, Hanting Chen, Yehui Tang, Tianyu Guo, Kai Han, Ying Nie, Xutao Wang, Hailin Hu, Zheyuan Bai, Yun Wang, Fangcheng Liu, Zhicheng Liu, Jianyuan Guo, Sinan Zeng, Yinchen Zhang, Qinghua Xu, Qun Liu, Jun Yao, Chao Xu, Dacheng Tao

The recent trend of large language models (LLMs) is to increase the scale of both model size (\aka the number of parameters) and dataset to achieve better generative ability, which is definitely proved by a lot of work such as the famous GPT and Llama. However, large models often involve massive computational costs, and practical applications cannot afford such high prices. However, the method of constructing a strong model architecture for LLMs is rarely discussed. We first analyze the state-of-the-art language model architectures and observe the feature collapse problem. Based on the theoretical analysis, we propose that the nonlinearity is also very important for language models, which is usually studied in convolutional neural networks for vision tasks. The series informed activation function is then introduced with tiny calculations that can be ignored, and an augmented shortcut is further used to enhance the model nonlinearity. We then demonstrate that the proposed approach is significantly effective for enhancing the model nonlinearity through carefully designed ablations; thus, we present a new efficient model architecture for establishing modern, namely, PanGu-$\pi$. Experiments are then conducted using the same dataset and training strategy to compare PanGu-$\pi$ with state-of-the-art LLMs. The results show that PanGu-$\pi$-7B can achieve a comparable performance to that of benchmarks with about 10\% inference speed-up, and PanGu-$\pi$-1B can achieve state-of-the-art performance in terms of accuracy and efficiency. In addition, we have deployed PanGu-$\pi$-7B in the high-value domains of finance and law, developing an LLM named YunShan for practical application. The results show that YunShan can surpass other models with similar scales on benchmarks.

http://arxiv.org/abs/2312.17276v1

cs.CL, cs.LG

Task Contamination: Language Models May Not Be Few-Shot Anymore

Changmao Li, Jeffrey Flanigan

Large language models (LLMs) offer impressive performance in various zero-shot and few-shot tasks. However, their success in zero-shot and few-shot settings may be affected by task contamination, a potential limitation that has not been thoroughly examined. This paper investigates how zero-shot and few-shot performance of LLMs has changed chronologically over time. Utilizing GPT-3 series models and several other recent open-sourced LLMs, and controlling for dataset difficulty, we find that on datasets released before the LLM training data creation date, LLMs perform surprisingly better than on datasets released after. This strongly indicates that, for many LLMs, there exists task contamination on zero-shot and few-shot evaluation for datasets released prior to the LLMs' training data creation date. Additionally, we utilize training data inspection, task example extraction, and a membership inference attack, which reveal further evidence of task contamination. Importantly, we find that for classification tasks with no possibility of task contamination, LLMs rarely demonstrate statistically significant improvements over simple majority baselines, in both zero and few-shot settings.

http://arxiv.org/abs/2312.16337v1

cs.CL, I.2.7

Understanding Before Recommendation: Semantic Aspect-Aware Review Exploitation via Large Language Models

Fan Liu, Yaqi Liu, Zhiyong Cheng, Liqiang Nie, Mohan Kankanhalli

Recommendation systems harness user-item interactions like clicks and reviews to learn their representations. Previous studies improve recommendation accuracy and interpretability by modeling user preferences across various aspects and intents. However, the aspects and intents are inferred directly from user reviews or behavior patterns, suffering from the data noise and the data sparsity problem. Furthermore, it is difficult to understand the reasons behind recommendations due to the challenges of interpreting implicit aspects and intents. Inspired by the deep semantic understanding offered by large language models (LLMs), we introduce a chain-based prompting approach to uncover semantic aspect-aware interactions, which provide clearer insights into user behaviors at a fine-grained semantic level. To incorporate the abundant interactions of various aspects, we propose the simple yet effective Semantic Aspect-based Graph Convolution Network (short for SAGCN). By performing graph convolutions on multiple semantic aspect graphs, SAGCN efficiently combines embeddings across multiple semantic aspects for final user and item representations. The effectiveness of the SAGCN was evaluated on three publicly available datasets through extensive experiments, which revealed that it outperforms all other competitors. Furthermore, interpretability analysis experiments were conducted to demonstrate the interpretability of incorporating semantic aspects into the model.

http://arxiv.org/abs/2312.16275v1

cs.IR, cs.MM, H.3.3

A Prompt Learning Framework for Source Code Summarization

Weisong Sun, Chunrong Fang, Yudu You, Yuchen Chen, Yi Liu, Chong Wang, Jian Zhang, Quanjun Zhang, Hanwei Qian, Wei Zhao, Yang Liu, Zhenyu Chen

(Source) code summarization is the task of automatically generating natural language summaries for given code snippets. Such summaries play a key role in helping developers understand and maintain source code. Recently, with the successful application of large language models (LLMs) in numerous fields, software engineering researchers have also attempted to adapt LLMs to solve code summarization tasks. The main adaptation schemes include instruction prompting and task-oriented fine-tuning. However, instruction prompting involves designing crafted prompts for zero-shot learning or selecting appropriate samples for few-shot learning and requires users to have professional domain knowledge, while task-oriented fine-tuning requires high training costs. In this paper, we propose a novel prompt learning framework for code summarization called PromptCS. PromptCS trains a prompt agent that can generate continuous prompts to unleash the potential for LLMs in code summarization. Compared to the human-written discrete prompt, the continuous prompts are produced under the guidance of LLMs and are therefore easier to understand by LLMs. PromptCS freezes the parameters of LLMs when training the prompt agent, which can greatly reduce the requirements for training resources. We evaluate PromptCS on the CodeSearchNet dataset involving multiple programming languages. The results show that PromptCS significantly outperforms instruction prompting schemes on all four widely used metrics. In some base LLMs, e.g., CodeGen-Multi-2B and StarCoderBase-1B and -3B, PromptCS even outperforms the task-oriented fine-tuning scheme. More importantly, the training efficiency of PromptCS is faster than the task-oriented fine-tuning scheme, with a more pronounced advantage on larger LLMs. The results of the human evaluation demonstrate that PromptCS can generate more good summaries compared to baselines.

http://arxiv.org/abs/2312.16066v1

cs.SE, cs.AI, 68-04, 68T30, D.2.3; I.2.2; I.2.4

LLMLight: Large Language Models as Traffic Signal Control Agents

Siqi Lai, Zhao Xu, Weijia Zhang, Hao Liu, Hui Xiong

Traffic Signal Control (TSC) is a crucial component in urban traffic management, aiming to optimize road network efficiency and reduce congestion. Traditional methods in TSC, primarily based on transportation engineering and reinforcement learning (RL), often exhibit limitations in generalization across varied traffic scenarios and lack interpretability. This paper presents LLMLight, a novel framework employing Large Language Models (LLMs) as decision-making agents for TSC. Specifically, the framework begins by instructing the LLM with a knowledgeable prompt detailing real-time traffic conditions. Leveraging the advanced generalization capabilities of LLMs, LLMLight engages a reasoning and decision-making process akin to human intuition for effective traffic control. Moreover, we build LightGPT, a specialized backbone LLM tailored for TSC tasks. By learning nuanced traffic patterns and control strategies, LightGPT enhances the LLMLight framework cost-effectively. Extensive experiments on nine real-world and synthetic datasets showcase the remarkable effectiveness, generalization ability, and interpretability of LLMLight against nine transportation-based and RL-based baselines.

http://arxiv.org/abs/2312.16044v4

cs.AI

RecRanker: Instruction Tuning Large Language Model as Ranker for Top-k Recommendation

Sichun Luo, Bowei He, Haohan Zhao, Wei Shao, Yanlin Qi, Yinya Huang, Aojun Zhou, Yuxuan Yao, Zongpeng Li, Yuanzhang Xiao, Mingjie Zhan, Linqi Song

Large Language Models (LLMs) have demonstrated remarkable capabilities and have been extensively deployed across various domains, including recommender systems. Prior research has employed specialized \textit{prompts} to leverage the in-context learning capabilities of LLMs for recommendation purposes. More recent studies have utilized instruction tuning techniques to align LLMs with human preferences, promising more effective recommendations. However, existing methods suffer from several limitations. The full potential of LLMs is not fully elicited due to low-quality tuning data and the overlooked integration of conventional recommender signals. Furthermore, LLMs may generate inconsistent responses for different ranking tasks in the recommendation, potentially leading to unreliable results. In this paper, we introduce \textbf{RecRanker}, tailored for instruction tuning LLMs to serve as the \textbf{Ranker} for top-\textit{k} \textbf{Rec}ommendations. Specifically, we introduce an adaptive sampling module for sampling high-quality, representative, and diverse training data. To enhance the prompt, we introduce a position shifting strategy to mitigate position bias and augment the prompt with auxiliary information from conventional recommendation models, thereby enriching the contextual understanding of the LLM. Subsequently, we utilize the sampled data to assemble an instruction-tuning dataset with the augmented prompts comprising three distinct ranking tasks: pointwise, pairwise, and listwise rankings. We further propose a hybrid ranking method to enhance the model performance by ensembling these ranking tasks. Our empirical evaluations demonstrate the effectiveness of our proposed RecRanker in both direct and sequential recommendation scenarios.

http://arxiv.org/abs/2312.16018v3

cs.IR

Towards Probing Contact Center Large Language Models

Varun Nathan, Ayush Kumar, Digvijay Ingle, Jithendra Vepa

Fine-tuning large language models (LLMs) with domain-specific instructions has emerged as an effective method to enhance their domain-specific understanding. Yet, there is limited work that examines the core characteristics acquired during this process. In this study, we benchmark the fundamental characteristics learned by contact-center (CC) specific instruction fine-tuned LLMs with out-of-the-box (OOB) LLMs via probing tasks encompassing conversational, channel, and automatic speech recognition (ASR) properties. We explore different LLM architectures (Flan-T5 and Llama), sizes (3B, 7B, 11B, 13B), and fine-tuning paradigms (full fine-tuning vs PEFT). Our findings reveal remarkable effectiveness of CC-LLMs on the in-domain downstream tasks, with improvement in response acceptability by over 48% compared to OOB-LLMs. Additionally, we compare the performance of OOB-LLMs and CC-LLMs on the widely used SentEval dataset, and assess their capabilities in terms of surface, syntactic, and semantic information through probing tasks. Intriguingly, we note a relatively consistent performance of probing classifiers on the set of probing tasks. Our observations indicate that CC-LLMs, while outperforming their out-of-the-box counterparts, exhibit a tendency to rely less on encoding surface, syntactic, and semantic properties, highlighting the intricate interplay between domain-specific adaptation and probing task performance opening up opportunities to explore behavior of fine-tuned language models in specialized contexts.

http://arxiv.org/abs/2312.15922v1

cs.CL

Supervised Knowledge Makes Large Language Models Better In-context Learners

Linyi Yang, Shuibai Zhang, Zhuohao Yu, Guangsheng Bao, Yidong Wang, Jindong Wang, Ruochen Xu, Wei Ye, Xing Xie, Weizhu Chen, Yue Zhang

Large Language Models (LLMs) exhibit emerging in-context learning abilities through prompt engineering. The recent progress in large-scale generative models has further expanded their use in real-world language applications. However, the critical challenge of improving the generalizability and factuality of LLMs in natural language understanding and question answering remains under-explored. While previous in-context learning research has focused on enhancing models to adhere to users' specific instructions and quality expectations, and to avoid undesired outputs, little to no work has explored the use of task-Specific fine-tuned Language Models (SLMs) to improve LLMs' in-context learning during the inference stage. Our primary contribution is the establishment of a simple yet effective framework that enhances the reliability of LLMs as it: 1) generalizes out-of-distribution data, 2) elucidates how LLMs benefit from discriminative models, and 3) minimizes hallucinations in generative tasks. Using our proposed plug-in method, enhanced versions of Llama 2 and ChatGPT surpass their original versions regarding generalizability and factuality. We offer a comprehensive suite of resources, including 16 curated datasets, prompts, model checkpoints, and LLM outputs across 9 distinct tasks. The code and data are released at: https://github.com/YangLinyi/Supervised-Knowledge-Makes-Large-Language-Models-Better-In-context-Learners. Our empirical analysis sheds light on the advantages of incorporating discriminative models into LLMs and highlights the potential of our methodology in fostering more reliable LLMs.

http://arxiv.org/abs/2312.15918v2

cs.CL, cs.AI

HyKGE: A Hypothesis Knowledge Graph Enhanced Framework for Accurate and Reliable Medical LLMs Responses

Xinke Jiang, Ruizhe Zhang, Yongxin Xu, Rihong Qiu, Yue Fang, Zhiyuan Wang, Jinyi Tang, Hongxin Ding, Xu Chu, Junfeng Zhao, Yasha Wang

In this paper, we investigate the retrieval-augmented generation (RAG) based on Knowledge Graphs (KGs) to improve the accuracy and reliability of Large Language Models (LLMs). Recent approaches suffer from insufficient and repetitive knowledge retrieval, tedious and time-consuming query parsing, and monotonous knowledge utilization. To this end, we develop a Hypothesis Knowledge Graph Enhanced (HyKGE) framework, which leverages LLMs' powerful reasoning capacity to compensate for the incompleteness of user queries, optimizes the interaction process with LLMs, and provides diverse retrieved knowledge. Specifically, HyKGE explores the zero-shot capability and the rich knowledge of LLMs with Hypothesis Outputs to extend feasible exploration directions in the KGs, as well as the carefully curated prompt to enhance the density and efficiency of LLMs' responses. Furthermore, we introduce the HO Fragment Granularity-aware Rerank Module to filter out noise while ensuring the balance between diversity and relevance in retrieved knowledge. Experiments on two Chinese medical multiple-choice question datasets and one Chinese open-domain medical Q&A dataset with two LLM turbos demonstrate the superiority of HyKGE in terms of accuracy and explainability.

http://arxiv.org/abs/2312.15883v2

cs.CL, cs.AI

Knowledge Distillation of LLM for Automatic Scoring of Science Education Assessments

Ehsan Latif, Luyang Fang, Ping Ma, Xiaoming Zhai

This study proposes a method for knowledge distillation (KD) of fine-tuned Large Language Models (LLMs) into smaller, more efficient, and accurate neural networks. We specifically target the challenge of deploying these models on resource-constrained devices. Our methodology involves training the smaller student model (Neural Network) using the prediction probabilities (as soft labels) of the LLM, which serves as a teacher model. This is achieved through a specialized loss function tailored to learn from the LLM's output probabilities, ensuring that the student model closely mimics the teacher's performance. To validate the performance of the KD approach, we utilized a large dataset, 7T, containing 6,684 student-written responses to science questions and three mathematical reasoning datasets with student-written responses graded by human experts. We compared accuracy with state-of-the-art (SOTA) distilled models, TinyBERT, and artificial neural network (ANN) models. Results have shown that the KD approach has 1% and 4% higher scoring accuracy than ANN and TinyBERT and comparable accuracy to the teacher model. Furthermore, the student model size is 0.02M, 10,000 times smaller in parameters and x10 faster in inferencing than the teacher model and TinyBERT, respectively. The significance of this research lies in its potential to make advanced AI technologies accessible in typical educational settings, particularly for automatic scoring.

http://arxiv.org/abs/2312.15842v2

cs.CL, cs.AI

SecQA: A Concise Question-Answering Dataset for Evaluating Large Language Models in Computer Security

Zefang Liu

In this paper, we introduce SecQA, a novel dataset tailored for evaluating the performance of Large Language Models (LLMs) in the domain of computer security. Utilizing multiple-choice questions generated by GPT-4 based on the "Computer Systems Security: Planning for Success" textbook, SecQA aims to assess LLMs' understanding and application of security principles. We detail the structure and intent of SecQA, which includes two versions of increasing complexity, to provide a concise evaluation across various difficulty levels. Additionally, we present an extensive evaluation of prominent LLMs, including GPT-3.5-Turbo, GPT-4, Llama-2, Vicuna, Mistral, and Zephyr models, using both 0-shot and 5-shot learning settings. Our results, encapsulated in the SecQA v1 and v2 datasets, highlight the varying capabilities and limitations of these models in the computer security context. This study not only offers insights into the current state of LLMs in understanding security-related content but also establishes SecQA as a benchmark for future advancements in this critical research area.

http://arxiv.org/abs/2312.15838v1

cs.CL, cs.CR

AHAM: Adapt, Help, Ask, Model -- Harvesting LLMs for literature mining

Boshko Koloski, Nada Lavrač, Bojan Cestnik, Senja Pollak, Blaž Škrlj, Andrej Kastrin

In an era marked by a rapid increase in scientific publications, researchers grapple with the challenge of keeping pace with field-specific advances. We present the `AHAM' methodology and a metric that guides the domain-specific \textbf{adapt}ation of the BERTopic topic modeling framework to improve scientific text analysis. By utilizing the LLaMa2 generative language model, we generate topic definitions via one-shot learning by crafting prompts with the \textbf{help} of domain experts to guide the LLM for literature mining by \textbf{asking} it to model the topic names. For inter-topic similarity evaluation, we leverage metrics from language generation and translation processes to assess lexical and semantic similarity of the generated topics. Our system aims to reduce both the ratio of outlier topics to the total number of topics and the similarity between topic definitions. The methodology has been assessed on a newly gathered corpus of scientific papers on literature-based discovery. Through rigorous evaluation by domain experts, AHAM has been validated as effective in uncovering intriguing and novel insights within broad research areas. We explore the impact of domain adaptation of sentence-transformers for the task of topic \textbf{model}ing using two datasets, each specialized to specific scientific domains within arXiv and medarxiv. We evaluate the impact of data size, the niche of adaptation, and the importance of domain adaptation. Our results suggest a strong interaction between domain adaptation and topic modeling precision in terms of outliers and topic definitions.

http://arxiv.org/abs/2312.15784v1

cs.CL, cs.AI

Large Language Models are Not Stable Recommender Systems

Tianhui Ma, Yuan Cheng, Hengshu Zhu, Hui Xiong

With the significant successes of large language models (LLMs) in many natural language processing tasks, there is growing interest among researchers in exploring LLMs for novel recommender systems. However, we have observed that directly using LLMs as a recommender system is usually unstable due to its inherent position bias. To this end, we introduce exploratory research and find consistent patterns of positional bias in LLMs that influence the performance of recommendation across a range of scenarios. Then, we propose a Bayesian probabilistic framework, STELLA (Stable LLM for Recommendation), which involves a two-stage pipeline. During the first probing stage, we identify patterns in a transition matrix using a probing detection dataset. And in the second recommendation stage, a Bayesian strategy is employed to adjust the biased output of LLMs with an entropy indicator. Therefore, our framework can capitalize on existing pattern information to calibrate instability of LLMs, and enhance recommendation performance. Finally, extensive experiments clearly validate the effectiveness of our framework.

http://arxiv.org/abs/2312.15746v1

cs.IR, cs.AI

IQAGPT: Image Quality Assessment with Vision-language and ChatGPT Models

Zhihao Chen, Bin Hu, Chuang Niu, Tao Chen, Yuxin Li, Hongming Shan, Ge Wang

Large language models (LLMs), such as ChatGPT, have demonstrated impressive capabilities in various tasks and attracted an increasing interest as a natural language interface across many domains. Recently, large vision-language models (VLMs) like BLIP-2 and GPT-4 have been intensively investigated, which learn rich vision-language correlation from image-text pairs. However, despite these developments, the application of LLMs and VLMs in image quality assessment (IQA), particularly in medical imaging, remains to be explored, which is valuable for objective performance evaluation and potential supplement or even replacement of radiologists' opinions. To this end, this paper introduces IQAGPT, an innovative image quality assessment system integrating an image quality captioning VLM with ChatGPT for generating quality scores and textual reports. First, we build a CT-IQA dataset for training and evaluation, comprising 1,000 CT slices with diverse quality levels professionally annotated. To better leverage the capabilities of LLMs, we convert annotated quality scores into semantically rich text descriptions using a prompt template. Second, we fine-tune the image quality captioning VLM on the CT-IQA dataset to generate quality descriptions. The captioning model fuses the image and text features through cross-modal attention. Third, based on the quality descriptions, users can talk with ChatGPT to rate image quality scores or produce a radiological quality report. Our preliminary results demonstrate the feasibility of assessing image quality with large models. Remarkably, our IQAGPT outperforms GPT-4 and CLIP-IQA, as well as the multi-task classification and regression models that solely rely on images.

http://arxiv.org/abs/2312.15663v1

cs.CV, cs.AI

Unlocking the Potential of Large Language Models for Explainable Recommendations

Yucong Luo, Mingyue Cheng, Hao Zhang, Junyu Lu, Qi Liu, Enhong Chen

Generating user-friendly explanations regarding why an item is recommended has become increasingly common, largely due to advances in language generation technology, which can enhance user trust and facilitate more informed decision-making when using online services. However, existing explainable recommendation systems focus on using small-size language models. It remains uncertain what impact replacing the explanation generator with the recently emerging large language models (LLMs) would have. Can we expect unprecedented results? In this study, we propose LLMXRec, a simple yet effective two-stage explainable recommendation framework aimed at further boosting the explanation quality by employing LLMs. Unlike most existing LLM-based recommendation works, a key characteristic of LLMXRec is its emphasis on the close collaboration between previous recommender models and LLM-based explanation generators. Specifically, by adopting several key fine-tuning techniques, including parameter-efficient instructing tuning and personalized prompt techniques, controllable and fluent explanations can be well generated to achieve the goal of explanation recommendation. Most notably, we provide three different perspectives to evaluate the effectiveness of the explanations. Finally, we conduct extensive experiments over several benchmark recommender models and publicly available datasets. The experimental results not only yield positive results in terms of effectiveness and efficiency but also uncover some previously unknown outcomes. To facilitate further explorations in this area, the full code and detailed original results are open-sourced at https://github.com/GodFire66666/LLM_rec_explanation/.

http://arxiv.org/abs/2312.15661v3

cs.IR, cs.AI

Chatbot is Not All You Need: Information-rich Prompting for More Realistic Responses

Seokhoon Jeong, Assentay Makhmud

Recent Large Language Models (LLMs) have shown remarkable capabilities in mimicking fictional characters or real humans in conversational settings. However, the realism and consistency of these responses can be further enhanced by providing richer information of the agent being mimicked. In this paper, we propose a novel approach to generate more realistic and consistent responses from LLMs, leveraging five senses, attributes, emotional states, relationship with the interlocutor, and memories. By incorporating these factors, we aim to increase the LLM's capacity for generating natural and realistic reactions in conversational exchanges. Through our research, we expect to contribute to the development of LLMs that demonstrate improved capabilities in mimicking fictional characters. We release a new benchmark dataset and all our codes, prompts, and sample results on our Github: https://github.com/srafsasm/InfoRichBot

http://arxiv.org/abs/2312.16233v1

cs.CL