Datasets:

bunkalab

/

arxiv_datasets_llm

like 0

LLaGA: Large Language and Graph Assistant

Runjin Chen, Tong Zhao, Ajay Jaiswal, Neil Shah, Zhangyang Wang

Graph Neural Networks (GNNs) have empowered the advance in graph-structured data analysis. Recently, the rise of Large Language Models (LLMs) like GPT-4 has heralded a new era in deep learning. However, their application to graph data poses distinct challenges due to the inherent difficulty of translating graph structures to language. To this end, we introduce the Large Language and Graph Assistant (LLaGA), an innovative model that effectively integrates LLM capabilities to handle the complexities of graph-structured data. LLaGA retains the general-purpose nature of LLMs while adapting graph data into a format compatible with LLM input. LLaGA achieves this by reorganizing graph nodes to structure-aware sequences and then mapping these into the token embedding space through a versatile projector. LLaGA excels in versatility, generalizability and interpretability, allowing it to perform consistently well across different datasets and tasks, extend its ability to unseen datasets or tasks, and provide explanations for graphs. Our extensive experiments across popular graph benchmarks show that LLaGA delivers outstanding performance across four datasets and three tasks using one single model, surpassing state-of-the-art graph models in both supervised and zero-shot scenarios. Our code is available at \url{https://github.com/VITA-Group/LLaGA}.

http://arxiv.org/abs/2402.08170v3

cs.LG, cs.AI

Investigating the Impact of Data Contamination of Large Language Models in Text-to-SQL Translation

Federico Ranaldi, Elena Sofia Ruzzetti, Dario Onorati, Leonardo Ranaldi, Cristina Giannone, Andrea Favalli, Raniero Romagnoli, Fabio Massimo Zanzotto

Understanding textual description to generate code seems to be an achieved capability of instruction-following Large Language Models (LLMs) in zero-shot scenario. However, there is a severe possibility that this translation ability may be influenced by having seen target textual descriptions and the related code. This effect is known as Data Contamination. In this study, we investigate the impact of Data Contamination on the performance of GPT-3.5 in the Text-to-SQL code-generating tasks. Hence, we introduce a novel method to detect Data Contamination in GPTs and examine GPT-3.5's Text-to-SQL performances using the known Spider Dataset and our new unfamiliar dataset Termite. Furthermore, we analyze GPT-3.5's efficacy on databases with modified information via an adversarial table disconnection (ATD) approach, complicating Text-to-SQL tasks by removing structural pieces of information from the database. Our results indicate a significant performance drop in GPT-3.5 on the unfamiliar Termite dataset, even with ATD modifications, highlighting the effect of Data Contamination on LLMs in Text-to-SQL translation tasks.

http://arxiv.org/abs/2402.08100v1

cs.CL, cs.LG

Walia-LLM: Enhancing Amharic-LLaMA by Integrating Task-Specific and Generative Datasets

Israel Abebe Azime, Atnafu Lambebo Tonja, Tadesse Destaw Belay, Mitiku Yohannes Fuge, Aman Kassahun Wassie, Eyasu Shiferaw Jada, Yonas Chanie, Walelign Tewabe Sewunetie, Seid Muhie Yimam

Large language models (LLMs) have received a lot of attention in natural language processing (NLP) research because of their exceptional performance in understanding and generating human languages. However, low-resource languages are left behind due to the unavailability of resources. In this work, we focus on enhancing the LLaMA-2-Amharic model by integrating task-specific and generative datasets to improve language model performance for Amharic. We compile an Amharic instruction fine-tuning dataset and fine-tuned LLaMA-2-Amharic model. The fine-tuned model shows promising results in different NLP tasks. We open-source our dataset creation pipeline, instruction datasets, trained models, and evaluation outputs to promote language-specific studies on these models.

http://arxiv.org/abs/2402.08015v5

cs.CL

Refined Direct Preference Optimization with Synthetic Data for Behavioral Alignment of LLMs

Víctor Gallego

In this paper, we introduce \emph{refined Direct Preference Optimization} (rDPO), a method for improving the behavioral alignment of Large Language Models (LLMs) without the need for human-annotated data. The method involves creating synthetic data using self-critique prompting by a teacher LLM and then utilising a generalized DPO loss function to distil to a student LLM. The loss function incorporates an additional external reward model to improve the quality of synthetic data, making rDPO robust to potential noise in the synthetic dataset. rDPO is shown to be effective in a diverse set of behavioural alignment tasks, such as improved safety, robustness against role-playing, and reduced sycophancy. Code to be released at https://github.com/vicgalle/refined-dpo.

http://arxiv.org/abs/2402.08005v1

cs.CL, cs.LG

Suppressing Pink Elephants with Direct Principle Feedback

Louis Castricato, Nathan Lile, Suraj Anand, Hailey Schoelkopf, Siddharth Verma, Stella Biderman

Existing methods for controlling language models, such as RLHF and Constitutional AI, involve determining which LLM behaviors are desirable and training them into a language model. However, in many cases, it is desirable for LLMs to be controllable at inference time, so that they can be used in multiple contexts with diverse needs. We illustrate this with the Pink Elephant Problem: instructing an LLM to avoid discussing a certain entity (a ``Pink Elephant''), and instead discuss a preferred entity (``Grey Elephant''). We apply a novel simplification of Constitutional AI, Direct Principle Feedback, which skips the ranking of responses and uses DPO directly on critiques and revisions. Our results show that after DPF fine-tuning on our synthetic Pink Elephants dataset, our 13B fine-tuned LLaMA 2 model significantly outperforms Llama-2-13B-Chat and a prompted baseline, and performs as well as GPT-4 in on our curated test set assessing the Pink Elephant Problem.

http://arxiv.org/abs/2402.07896v2

cs.CL

PoisonedRAG: Knowledge Poisoning Attacks to Retrieval-Augmented Generation of Large Language Models

Wei Zou, Runpeng Geng, Binghui Wang, Jinyuan Jia

Large language models (LLMs) have achieved remarkable success due to their exceptional generative capabilities. Despite their success, they also have inherent limitations such as a lack of up-to-date knowledge and hallucination. Retrieval-Augmented Generation (RAG) is a state-of-the-art technique to mitigate those limitations. In particular, given a question, RAG retrieves relevant knowledge from a knowledge database to augment the input of the LLM. For instance, the retrieved knowledge could be a set of top-k texts that are most semantically similar to the given question when the knowledge database contains millions of texts collected from Wikipedia. As a result, the LLM could utilize the retrieved knowledge as the context to generate an answer for the given question. Existing studies mainly focus on improving the accuracy or efficiency of RAG, leaving its security largely unexplored. We aim to bridge the gap in this work. Particularly, we propose PoisonedRAG , a set of knowledge poisoning attacks to RAG, where an attacker could inject a few poisoned texts into the knowledge database such that the LLM generates an attacker-chosen target answer for an attacker-chosen target question. We formulate knowledge poisoning attacks as an optimization problem, whose solution is a set of poisoned texts. Depending on the background knowledge (e.g., black-box and white-box settings) of an attacker on the RAG, we propose two solutions to solve the optimization problem, respectively. Our results on multiple benchmark datasets and LLMs show our attacks could achieve 90% attack success rates when injecting 5 poisoned texts for each target question into a database with millions of texts. We also evaluate recent defenses and our results show they are insufficient to defend against our attacks, highlighting the need for new defenses.

http://arxiv.org/abs/2402.07867v1

cs.CR, cs.LG

Lissard: Long and Simple Sequential Reasoning Datasets

Mirelle Bueno, Roberto Lotufo, Rodrigo Nogueira

Language models are now capable of solving tasks that require dealing with long sequences consisting of hundreds of thousands of tokens. However, they often fail on tasks that require repetitive use of simple rules, even on sequences that are much shorter than those seen during training. For example, state-of-the-art LLMs can find common items in two lists with up to 20 items but fail when lists have 80 items. In this paper, we introduce Lissard, a benchmark comprising seven tasks whose goal is to assess the ability of models to process and generate wide-range sequence lengths, requiring repetitive procedural execution. Our evaluation of open-source (Mistral-7B and Mixtral-8x7B) and proprietary models (GPT-3.5 and GPT-4) show a consistent decline in performance across all models as the complexity of the sequence increases. The datasets and code are available at https://github.com/unicamp-dl/Lissard

http://arxiv.org/abs/2402.07859v2

cs.CL, cs.AI

Do Membership Inference Attacks Work on Large Language Models?

Michael Duan, Anshuman Suri, Niloofar Mireshghallah, Sewon Min, Weijia Shi, Luke Zettlemoyer, Yulia Tsvetkov, Yejin Choi, David Evans, Hannaneh Hajishirzi

Membership inference attacks (MIAs) attempt to predict whether a particular datapoint is a member of a target model's training data. Despite extensive research on traditional machine learning models, there has been limited work studying MIA on the pre-training data of large language models (LLMs). We perform a large-scale evaluation of MIAs over a suite of language models (LMs) trained on the Pile, ranging from 160M to 12B parameters. We find that MIAs barely outperform random guessing for most settings across varying LLM sizes and domains. Our further analyses reveal that this poor performance can be attributed to (1) the combination of a large dataset and few training iterations, and (2) an inherently fuzzy boundary between members and non-members. We identify specific settings where LLMs have been shown to be vulnerable to membership inference and show that the apparent success in such settings can be attributed to a distribution shift, such as when members and non-members are drawn from the seemingly identical domain but with different temporal ranges. We release our code and data as a unified benchmark package that includes all existing MIAs, supporting future work.

http://arxiv.org/abs/2402.07841v1

cs.CL

Aya Model: An Instruction Finetuned Open-Access Multilingual Language Model

Ahmet Üstün, Viraat Aryabumi, Zheng-Xin Yong, Wei-Yin Ko, Daniel D'souza, Gbemileke Onilude, Neel Bhandari, Shivalika Singh, Hui-Lee Ooi, Amr Kayid, Freddie Vargus, Phil Blunsom, Shayne Longpre, Niklas Muennighoff, Marzieh Fadaee, Julia Kreutzer, Sara Hooker

Recent breakthroughs in large language models (LLMs) have centered around a handful of data-rich languages. What does it take to broaden access to breakthroughs beyond first-class citizen languages? Our work introduces Aya, a massively multilingual generative language model that follows instructions in 101 languages of which over 50% are considered as lower-resourced. Aya outperforms mT0 and BLOOMZ on the majority of tasks while covering double the number of languages. We introduce extensive new evaluation suites that broaden the state-of-art for multilingual eval across 99 languages -- including discriminative and generative tasks, human evaluation, and simulated win rates that cover both held-out tasks and in-distribution performance. Furthermore, we conduct detailed investigations on the optimal finetuning mixture composition, data pruning, as well as the toxicity, bias, and safety of our models. We open-source our instruction datasets and our model at https://hf.co/CohereForAI/aya-101

http://arxiv.org/abs/2402.07827v1

cs.CL

Differentially Private Zeroth-Order Methods for Scalable Large Language Model Finetuning

Z Liu, J Lou, W Bao, Y Hu, B Li, Z Qin, K Ren

Fine-tuning on task-specific datasets is a widely-embraced paradigm of harnessing the powerful capability of pretrained LLMs for various downstream tasks. Due to the popularity of LLMs fine-tuning and its accompanying privacy concerns, differentially private (DP) fine-tuning of pretrained LLMs has been widely used to safeguarding the privacy of task-specific datasets. Lying at the design core of DP LLM fine-tuning methods is the satisfactory tradeoff among privacy, utility, and scalability. Most existing methods build upon the seminal work of DP-SGD. Despite pushing the scalability of DP-SGD to its limit, DP-SGD-based fine-tuning methods are unfortunately limited by the inherent inefficiency of SGD. In this paper, we investigate the potential of DP zeroth-order methods for LLM pretraining, which avoids the scalability bottleneck of SGD by approximating the gradient with the more efficient zeroth-order gradient. Rather than treating the zeroth-order method as a drop-in replacement for SGD, this paper presents a comprehensive study both theoretically and empirically. First, we propose the stagewise DP zeroth-order method (DP-ZOSO) that dynamically schedules key hyperparameters. This design is grounded on the synergy between DP random perturbation and the gradient approximation error of the zeroth-order method, and its effect on fine-tuning trajectory. We provide theoretical analysis for both proposed methods. We conduct extensive empirical analysis on both encoder-only masked language model and decoder-only autoregressive language model, achieving impressive results in terms of scalability and utility (compared with DPZero, DP-ZOPO improves 4.5% on SST-5, 5.5% on MNLI with RoBERTa-Large and 9.2% on CB, 3.9% on BoolQ with OPT-2.7B when $\epsilon=4$).

http://arxiv.org/abs/2402.07818v3

cs.LG, cs.AI, cs.CL

Empowering Federated Learning for Massive Models with NVIDIA FLARE

Holger R. Roth, Ziyue Xu, Yuan-Ting Hsieh, Adithya Renduchintala, Isaac Yang, Zhihong Zhang, Yuhong Wen, Sean Yang, Kevin Lu, Kristopher Kersten, Camir Ricketts, Daguang Xu, Chester Chen, Yan Cheng, Andrew Feng

In the ever-evolving landscape of artificial intelligence (AI) and large language models (LLMs), handling and leveraging data effectively has become a critical challenge. Most state-of-the-art machine learning algorithms are data-centric. However, as the lifeblood of model performance, necessary data cannot always be centralized due to various factors such as privacy, regulation, geopolitics, copyright issues, and the sheer effort required to move vast datasets. In this paper, we explore how federated learning enabled by NVIDIA FLARE can address these challenges with easy and scalable integration capabilities, enabling parameter-efficient and full supervised fine-tuning of LLMs for natural language processing and biopharmaceutical applications to enhance their accuracy and robustness.

http://arxiv.org/abs/2402.07792v1

cs.LG, cs.DC

TELLER: A Trustworthy Framework for Explainable, Generalizable and Controllable Fake News Detection

Hui Liu, Wenya Wang, Haoru Li, Haoliang Li

The proliferation of fake news has emerged as a severe societal problem, raising significant interest from industry and academia. While existing deep-learning based methods have made progress in detecting fake news accurately, their reliability may be compromised caused by the non-transparent reasoning processes, poor generalization abilities and inherent risks of integration with large language models (LLMs). To address this challenge, we propose {\methodname}, a novel framework for trustworthy fake news detection that prioritizes explainability, generalizability and controllability of models. This is achieved via a dual-system framework that integrates cognition and decision systems, adhering to the principles above. The cognition system harnesses human expertise to generate logical predicates, which guide LLMs in generating human-readable logic atoms. Meanwhile, the decision system deduces generalizable logic rules to aggregate these atoms, enabling the identification of the truthfulness of the input news across diverse domains and enhancing transparency in the decision-making process. Finally, we present comprehensive evaluation results on four datasets, demonstrating the feasibility and trustworthiness of our proposed framework. Our implementation is available at \url{https://github.com/less-and-less-bugs/Trust_TELLER}.

http://arxiv.org/abs/2402.07776v1

cs.CL

CyberMetric: A Benchmark Dataset for Evaluating Large Language Models Knowledge in Cybersecurity

Norbert Tihanyi, Mohamed Amine Ferrag, Ridhi Jain, Merouane Debbah

Large Language Models (LLMs) excel across various domains, from computer vision to medical diagnostics. However, understanding the diverse landscape of cybersecurity, encompassing cryptography, reverse engineering, and managerial facets like risk assessment, presents a challenge, even for human experts. In this paper, we introduce CyberMetric, a benchmark dataset comprising 10,000 questions sourced from standards, certifications, research papers, books, and other publications in the cybersecurity domain. The questions are created through a collaborative process, i.e., merging expert knowledge with LLMs, including GPT-3.5 and Falcon-180B. Human experts spent over 200 hours verifying their accuracy and relevance. Beyond assessing LLMs' knowledge, the dataset's main goal is to facilitate a fair comparison between humans and different LLMs in cybersecurity. To achieve this, we carefully selected 80 questions covering a wide range of topics within cybersecurity and involved 30 participants of diverse expertise levels, facilitating a comprehensive comparison between human and machine intelligence in this area. The findings revealed that LLMs outperformed humans in almost every aspect of cybersecurity.

http://arxiv.org/abs/2402.07688v1

cs.AI, cs.CR

The Sound of Healthcare: Improving Medical Transcription ASR Accuracy with Large Language Models

Ayo Adedeji, Sarita Joshi, Brendan Doohan

In the rapidly evolving landscape of medical documentation, transcribing clinical dialogues accurately is increasingly paramount. This study explores the potential of Large Language Models (LLMs) to enhance the accuracy of Automatic Speech Recognition (ASR) systems in medical transcription. Utilizing the PriMock57 dataset, which encompasses a diverse range of primary care consultations, we apply advanced LLMs to refine ASR-generated transcripts. Our research is multifaceted, focusing on improvements in general Word Error Rate (WER), Medical Concept WER (MC-WER) for the accurate transcription of essential medical terms, and speaker diarization accuracy. Additionally, we assess the role of LLM post-processing in improving semantic textual similarity, thereby preserving the contextual integrity of clinical dialogues. Through a series of experiments, we compare the efficacy of zero-shot and Chain-of-Thought (CoT) prompting techniques in enhancing diarization and correction accuracy. Our findings demonstrate that LLMs, particularly through CoT prompting, not only improve the diarization accuracy of existing ASR systems but also achieve state-of-the-art performance in this domain. This improvement extends to more accurately capturing medical concepts and enhancing the overall semantic coherence of the transcribed dialogues. These findings illustrate the dual role of LLMs in augmenting ASR outputs and independently excelling in transcription tasks, holding significant promise for transforming medical ASR systems and leading to more accurate and reliable patient records in healthcare settings.

http://arxiv.org/abs/2402.07658v1

cs.CL, cs.SD, eess.AS

G-Retriever: Retrieval-Augmented Generation for Textual Graph Understanding and Question Answering

Xiaoxin He, Yijun Tian, Yifei Sun, Nitesh V. Chawla, Thomas Laurent, Yann LeCun, Xavier Bresson, Bryan Hooi

Given a graph with textual attributes, we enable users to `chat with their graph': that is, to ask questions about the graph using a conversational interface. In response to a user's questions, our method provides textual replies and highlights the relevant parts of the graph. While existing works integrate large language models (LLMs) and graph neural networks (GNNs) in various ways, they mostly focus on either conventional graph tasks (such as node, edge, and graph classification), or on answering simple graph queries on small or synthetic graphs. In contrast, we develop a flexible question-answering framework targeting real-world textual graphs, applicable to multiple applications including scene graph understanding, common sense reasoning, and knowledge graph reasoning. Toward this goal, we first develop our Graph Question Answering (GraphQA) benchmark with data collected from different tasks. Then, we propose our G-Retriever approach, which integrates the strengths of GNNs, LLMs, and Retrieval-Augmented Generation (RAG), and can be fine-tuned to enhance graph understanding via soft prompting. To resist hallucination and to allow for textual graphs that greatly exceed the LLM's context window size, G-Retriever performs RAG over a graph by formulating this task as a Prize-Collecting Steiner Tree optimization problem. Empirical evaluations show that our method outperforms baselines on textual graph tasks from multiple domains, scales well with larger graph sizes, and resists hallucination. (Our codes and datasets are available at: https://github.com/XiaoxinHe/G-Retriever.)

http://arxiv.org/abs/2402.07630v2

cs.LG

Food Recommendation as Language Processing (F-RLP): A Personalized and Contextual Paradigm

Ali Rostami, Ramesh Jain, Amir M. Rahmani

State-of-the-art rule-based and classification-based food recommendation systems face significant challenges in becoming practical and useful. This difficulty arises primarily because most machine learning models struggle with problems characterized by an almost infinite number of classes and a limited number of samples within an unbalanced dataset. Conversely, the emergence of Large Language Models (LLMs) as recommendation engines offers a promising avenue. However, a general-purpose Recommendation as Language Processing (RLP) approach lacks the critical components necessary for effective food recommendations. To address this gap, we introduce Food Recommendation as Language Processing (F-RLP), a novel framework that offers a food-specific, tailored infrastructure. F-RLP leverages the capabilities of LLMs to maximize their potential, thereby paving the way for more accurate, personalized food recommendations.

http://arxiv.org/abs/2402.07477v2

cs.AI

Large Language Models are Few-shot Generators: Proposing Hybrid Prompt Algorithm To Generate Webshell Escape Samples

Mingrui Ma, Lansheng Han, Chunjie Zhou

The frequent occurrence of cyber-attacks has made webshell attacks and defense gradually become a research hotspot in the field of network security. However, the lack of publicly available benchmark datasets and the over-reliance on manually defined rules for webshell escape sample generation have slowed down the progress of research related to webshell escape sample generation strategies and artificial intelligence-based webshell detection algorithms. To address the drawbacks of weak webshell sample escape capabilities, the lack of webshell datasets with complex malicious features, and to promote the development of webshell detection technology, we propose the Hybrid Prompt algorithm for webshell escape sample generation with the help of large language models. As a prompt algorithm specifically developed for webshell sample generation, the Hybrid Prompt algorithm not only combines various prompt ideas including Chain of Thought, Tree of Thought, but also incorporates various components such as webshell hierarchical module and few-shot example to facilitate the LLM in learning and reasoning webshell escape strategies. Experimental results show that the Hybrid Prompt algorithm can work with multiple LLMs with excellent code reasoning ability to generate high-quality webshell samples with high Escape Rate (88.61% with GPT-4 model on VIRUSTOTAL detection engine) and Survival Rate (54.98% with GPT-4 model).

http://arxiv.org/abs/2402.07408v1

cs.CR, cs.AI

Dólares or Dollars? Unraveling the Bilingual Prowess of Financial LLMs Between Spanish and English

Xiao Zhang, Ruoyu Xiang, Chenhan Yuan, Duanyu Feng, Weiguang Han, Alejandro Lopez-Lira, Xiao-Yang Liu, Sophia Ananiadou, Min Peng, Jimin Huang, Qianqian Xie

Despite Spanish's pivotal role in the global finance industry, a pronounced gap exists in Spanish financial natural language processing (NLP) and application studies compared to English, especially in the era of large language models (LLMs). To bridge this gap, we unveil Tois\'on de Oro, the first bilingual framework that establishes instruction datasets, finetuned LLMs, and evaluation benchmark for financial LLMs in Spanish joint with English. We construct a rigorously curated bilingual instruction dataset including over 144K Spanish and English samples from 15 datasets covering 7 tasks. Harnessing this, we introduce FinMA-ES, an LLM designed for bilingual financial applications. We evaluate our model and existing LLMs using FLARE-ES, the first comprehensive bilingual evaluation benchmark with 21 datasets covering 9 tasks. The FLARE-ES benchmark results reveal a significant multilingual performance gap and bias in existing LLMs. FinMA-ES models surpass SOTA LLMs such as GPT-4 in Spanish financial tasks, due to strategic instruction tuning and leveraging data from diverse linguistic resources, highlighting the positive impact of cross-linguistic transfer. All our datasets, models, and benchmarks have been released.

http://arxiv.org/abs/2402.07405v1

cs.CL

Differentially Private Training of Mixture of Experts Models

Pierre Tholoniat, Huseyin A. Inan, Janardhan Kulkarni, Robert Sim

This position paper investigates the integration of Differential Privacy (DP) in the training of Mixture of Experts (MoE) models within the field of natural language processing. As Large Language Models (LLMs) scale to billions of parameters, leveraging expansive datasets, they exhibit enhanced linguistic capabilities and emergent abilities. However, this growth raises significant computational and privacy concerns. Our study addresses these issues by exploring the potential of MoE models, known for their computational efficiency, and the application of DP, a standard for privacy preservation. We present the first known attempt to train MoE models under the constraints of DP, addressing the unique challenges posed by their architecture and the complexities of DP integration. Our initial experimental studies demonstrate that MoE models can be effectively trained with DP, achieving performance that is competitive with their non-private counterparts. This initial study aims to provide valuable insights and ignite further research in the domain of privacy-preserving MoE models, softly laying the groundwork for prospective developments in this evolving field.

http://arxiv.org/abs/2402.07334v1

cs.CR, cs.LG

Online Iterative Reinforcement Learning from Human Feedback with General Preference Model

Chenlu Ye, Wei Xiong, Yuheng Zhang, Nan Jiang, Tong Zhang

We study Reinforcement Learning from Human Feedback (RLHF) under a general preference oracle. In particular, we do not assume that there exists a reward function and the preference signal is drawn from the Bradley-Terry model as most of the prior works do. We consider a standard mathematical formulation, the reverse-KL regularized minimax game between two LLMs for RLHF under general preference oracle. The learning objective of this formulation is to find a policy so that it is consistently preferred by the KL-regularized preference oracle over any competing LLMs. We show that this framework is strictly more general than the reward-based one, and propose sample-efficient algorithms for both the offline learning from a pre-collected preference dataset and online learning where we can query the preference oracle along the way of training. Empirical studies verify the effectiveness of the proposed framework.

http://arxiv.org/abs/2402.07314v2

cs.LG, stat.ML

Insights into Natural Language Database Query Errors: From Attention Misalignment to User Handling Strategies

Zheng Ning, Yuan Tian, Zheng Zhang, Tianyi Zhang, Toby Li

Querying structured databases with natural language (NL2SQL) has remained a difficult problem for years. Recently, the advancement of machine learning (ML), natural language processing (NLP), and large language models (LLM) have led to significant improvements in performance, with the best model achieving ~85% percent accuracy on the benchmark Spider dataset. However, there is a lack of a systematic understanding of the types, causes, and effectiveness of error-handling mechanisms of errors for erroneous queries nowadays. To bridge the gap, a taxonomy of errors made by four representative NL2SQL models was built in this work, along with an in-depth analysis of the errors. Second, the causes of model errors were explored by analyzing the model-human attention alignment to the natural language query. Last, a within-subjects user study with 26 participants was conducted to investigate the effectiveness of three interactive error-handling mechanisms in NL2SQL. Findings from this paper shed light on the design of model structure and error discovery and repair strategies for natural language data query interfaces in the future.

http://arxiv.org/abs/2402.07304v1

cs.HC

Previously on the Stories: Recap Snippet Identification for Story Reading

Jiangnan Li, Qiujing Wang, Liyan Xu, Wenjie Pang, Mo Yu, Zheng Lin, Weiping Wang, Jie Zhou

Similar to the "previously-on" scenes in TV shows, recaps can help book reading by recalling the readers' memory about the important elements in previous texts to better understand the ongoing plot. Despite its usefulness, this application has not been well studied in the NLP community. We propose the first benchmark on this useful task called Recap Snippet Identification with a hand-crafted evaluation dataset. Our experiments show that the proposed task is challenging to PLMs, LLMs, and proposed methods as the task requires a deep understanding of the plot correlation between snippets.

http://arxiv.org/abs/2402.07271v1

cs.CL

Open-ended VQA benchmarking of Vision-Language models by exploiting Classification datasets and their semantic hierarchy

Simon Ging, María A. Bravo, Thomas Brox

The evaluation of text-generative vision-language models is a challenging yet crucial endeavor. By addressing the limitations of existing Visual Question Answering (VQA) benchmarks and proposing innovative evaluation methodologies, our research seeks to advance our understanding of these models' capabilities. We propose a novel VQA benchmark based on well-known visual classification datasets which allows a granular evaluation of text-generative vision-language models and their comparison with discriminative vision-language models. To improve the assessment of coarse answers on fine-grained classification tasks, we suggest using the semantic hierarchy of the label space to ask automatically generated follow-up questions about the ground-truth category. Finally, we compare traditional NLP and LLM-based metrics for the problem of evaluating model predictions given ground-truth answers. We perform a human evaluation study upon which we base our decision on the final metric. We apply our benchmark to a suite of vision-language models and show a detailed comparison of their abilities on object, action, and attribute classification. Our contributions aim to lay the foundation for more precise and meaningful assessments, facilitating targeted progress in the exciting field of vision-language modeling.

http://arxiv.org/abs/2402.07270v2

cs.CV, cs.CL, cs.LG

CPSDBench: A Large Language Model Evaluation Benchmark and Baseline for Chinese Public Security Domain

Xin Tong, Bo Jin, Zhi Lin, Binjun Wang, Ting Yu, Qiang Cheng

Large Language Models (LLMs) have demonstrated significant potential and effectiveness across multiple application domains. To assess the performance of mainstream LLMs in public security tasks, this study aims to construct a specialized evaluation benchmark tailored to the Chinese public security domain--CPSDbench. CPSDbench integrates datasets related to public security collected from real-world scenarios, supporting a comprehensive assessment of LLMs across four key dimensions: text classification, information extraction, question answering, and text generation. Furthermore, this study introduces a set of innovative evaluation metrics designed to more precisely quantify the efficacy of LLMs in executing tasks related to public security. Through the in-depth analysis and evaluation conducted in this research, we not only enhance our understanding of the performance strengths and limitations of existing models in addressing public security issues but also provide references for the future development of more accurate and customized LLM models targeted at applications in this field.

http://arxiv.org/abs/2402.07234v3

cs.AI

Beware of Words: Evaluating the Lexical Richness of Conversational Large Language Models

Gonzalo Martínez, José Alberto Hernández, Javier Conde, Pedro Reviriego, Elena Merino

The performance of conversational Large Language Models (LLMs) in general, and of ChatGPT in particular, is currently being evaluated on many different tasks, from logical reasoning or maths to answering questions on a myriad of topics. Instead, much less attention is being devoted to the study of the linguistic features of the texts generated by these LLMs. This is surprising since LLMs are models for language, and understanding how they use the language is important. Indeed, conversational LLMs are poised to have a significant impact on the evolution of languages as they may eventually dominate the creation of new text. This means that for example, if conversational LLMs do not use a word it may become less and less frequent and eventually stop being used altogether. Therefore, evaluating the linguistic features of the text they produce and how those depend on the model parameters is the first step toward understanding the potential impact of conversational LLMs on the evolution of languages. In this paper, we consider the evaluation of the lexical richness of the text generated by LLMs and how it depends on the model parameters. A methodology is presented and used to conduct a comprehensive evaluation of lexical richness using ChatGPT as a case study. The results show how lexical richness depends on the version of ChatGPT and some of its parameters, such as the presence penalty, or on the role assigned to the model. The dataset and tools used in our analysis are released under open licenses with the goal of drawing the much-needed attention to the evaluation of the linguistic features of LLM-generated text.

http://arxiv.org/abs/2402.15518v1

cs.CL

Synergizing Spatial Optimization with Large Language Models for Open-Domain Urban Itinerary Planning

Yihong Tang, Zhaokai Wang, Ao Qu, Yihao Yan, Kebing Hou, Dingyi Zhuang, Xiaotong Guo, Jinhua Zhao, Zhan Zhao, Wei Ma

In this paper, we for the first time propose the task of Open-domain Urban Itinerary Planning (OUIP) for citywalk, which directly generates itineraries based on users' requests described in natural language. OUIP is different from conventional itinerary planning, which limits users from expressing more detailed needs and hinders true personalization. Recently, large language models (LLMs) have shown potential in handling diverse tasks. However, due to non-real-time information, incomplete knowledge, and insufficient spatial awareness, they are unable to independently deliver a satisfactory user experience in OUIP. Given this, we present ItiNera, an OUIP system that synergizes spatial optimization with Large Language Models (LLMs) to provide services that customize urban itineraries based on users' needs. Specifically, we develop an LLM-based pipeline for extracting and updating POI features to create a user-owned personalized POI database. For each user request, we leverage LLM in cooperation with an embedding-based module for retrieving candidate POIs from the user's POI database. Then, a spatial optimization module is used to order these POIs, followed by LLM crafting a personalized, spatially coherent itinerary. To the best of our knowledge, this study marks the first integration of LLMs to innovate itinerary planning solutions. Extensive experiments on offline datasets and online subjective evaluation have demonstrated the capacities of our system to deliver more responsive and spatially coherent itineraries than current LLM-based solutions. Our system has been deployed in production at the TuTu online travel service and has attracted thousands of users for their urban travel planning.

http://arxiv.org/abs/2402.07204v1

cs.AI, cs.CL, cs.LG

Unprecedented Code Change Automation: The Fusion of LLMs and Transformation by Example

Malinda Dilhara, Abhiram Bellur, Timofey Bryksin, Danny Dig

Software developers often repeat code changes, known as "code change patterns" (CPATs), within and across projects. Automating these CPATs accelerates development, but current Transformation by Example (TBE) techniques are limited by the input examples' quality and quantity, missing variations with different syntax or flow yet semantically similar. Large Language Models (LLMs), trained on vast code datasets, can overcome these limitations by generating semantically equivalent, unseen CPAT variants, enhancing TBE effectiveness. We identified best practices for using LLMs to generate code variants meeting criteria of correctness, usefulness, and applicability. Implementing these in PyCraft, combining static and dynamic analysis with LLMs, we achieved an F-measure of 96.6% in identifying correct variants, expanding inputs by 58x on average, and automating changes to increase target codes by up to 39x. Patches from PyCraft were submitted to projects like microsoft/DeepSpeed and IBM/inFairness, with an 83% acceptance rate, validating our approach's usefulness.

http://arxiv.org/abs/2402.07138v2

cs.SE

Generalizing Conversational Dense Retrieval via LLM-Cognition Data Augmentation

Haonan Chen, Zhicheng Dou, Kelong Mao, Jiongnan Liu, Ziliang Zhao

Conversational search utilizes muli-turn natural language contexts to retrieve relevant passages. Existing conversational dense retrieval models mostly view a conversation as a fixed sequence of questions and responses, overlooking the severe data sparsity problem -- that is, users can perform a conversation in various ways, and these alternate conversations are unrecorded. Consequently, they often struggle to generalize to diverse conversations in real-world scenarios. In this work, we propose a framework for generalizing Conversational dense retrieval via LLM-cognition data Augmentation (ConvAug). ConvAug first generates multi-level augmented conversations to capture the diverse nature of conversational contexts. Inspired by human cognition, we devise a cognition-aware process to mitigate the generation of false positives, false negatives, and hallucinations. Moreover, we develop a difficulty-adaptive sample filter that selects challenging samples for complex conversations, thereby giving the model a larger learning space. A contrastive learning objective is then employed to train a better conversational context encoder. Extensive experiments conducted on four public datasets, under both normal and zero-shot settings, demonstrate the effectiveness, generalizability, and applicability of ConvAug.

http://arxiv.org/abs/2402.07092v2

cs.CL, cs.IR

Gemini Goes to Med School: Exploring the Capabilities of Multimodal Large Language Models on Medical Challenge Problems & Hallucinations

Ankit Pal, Malaikannan Sankarasubbu

Large language models have the potential to be valuable in the healthcare industry, but it's crucial to verify their safety and effectiveness through rigorous evaluation. For this purpose, we comprehensively evaluated both open-source LLMs and Google's new multimodal LLM called Gemini across Medical reasoning, hallucination detection, and Medical Visual Question Answering tasks. While Gemini showed competence, it lagged behind state-of-the-art models like MedPaLM 2 and GPT-4 in diagnostic accuracy. Additionally, Gemini achieved an accuracy of 61.45\% on the medical VQA dataset, significantly lower than GPT-4V's score of 88\%. Our analysis revealed that Gemini is highly susceptible to hallucinations, overconfidence, and knowledge gaps, which indicate risks if deployed uncritically. We also performed a detailed analysis by medical subject and test type, providing actionable feedback for developers and clinicians. To mitigate risks, we applied prompting strategies that improved performance. Additionally, we facilitated future research and development by releasing a Python module for medical LLM evaluation and establishing a dedicated leaderboard on Hugging Face for medical domain LLMs. Python module can be found at https://github.com/promptslab/RosettaEval

http://arxiv.org/abs/2402.07023v1

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

OpenFedLLM: Training Large Language Models on Decentralized Private Data via Federated Learning

Rui Ye, Wenhao Wang, Jingyi Chai, Dihan Li, Zexi Li, Yinda Xu, Yaxin Du, Yanfeng Wang, Siheng Chen

Trained on massive publicly available data, large language models (LLMs) have demonstrated tremendous success across various fields. While more data contributes to better performance, a disconcerting reality is that high-quality public data will be exhausted in a few years. In this paper, we offer a potential next step for contemporary LLMs: collaborative and privacy-preserving LLM training on the underutilized distributed private data via federated learning (FL), where multiple data owners collaboratively train a shared model without transmitting raw data. To achieve this, we build a concise, integrated, and research-friendly framework/codebase, named OpenFedLLM. It covers federated instruction tuning for enhancing instruction-following capability, federated value alignment for aligning with human values, and 7 representative FL algorithms. Besides, OpenFedLLM supports training on diverse domains, where we cover 8 training datasets; and provides comprehensive evaluations, where we cover 30+ evaluation metrics. Through extensive experiments, we observe that all FL algorithms outperform local training on training LLMs, demonstrating a clear performance improvement across a variety of settings. Notably, in a financial benchmark, Llama2-7B fine-tuned by applying any FL algorithm can outperform GPT-4 by a significant margin while the model obtained through individual training cannot, demonstrating strong motivation for clients to participate in FL. The code is available at https://github.com/rui-ye/OpenFedLLM.

http://arxiv.org/abs/2402.06954v1

cs.LG, cs.CL, cs.DC, cs.MA

Can LLMs Recognize Toxicity? Structured Toxicity Investigation Framework and Semantic-Based Metric

Hyukhun Koh, Dohyung Kim, Minwoo Lee, Kyomin Jung

In the pursuit of developing Large Language Models (LLMs) that adhere to societal standards, it is imperative to discern the existence of toxicity in the generated text. The majority of existing toxicity metrics rely on encoder models trained on specific toxicity datasets. However, these encoders are susceptible to out-of-distribution (OOD) problems and depend on the definition of toxicity assumed in a dataset. In this paper, we introduce an automatic robust metric grounded on LLMs to distinguish whether model responses are toxic. We start by analyzing the toxicity factors, followed by examining the intrinsic toxic attributes of LLMs to ascertain their suitability as evaluators. Subsequently, we evaluate our metric, LLMs As ToxiciTy Evaluators (LATTE), on evaluation datasets.The empirical results indicate outstanding performance in measuring toxicity, improving upon state-of-the-art metrics by 12 points in F1 score without training procedure. We also show that upstream toxicity has an influence on downstream metrics.

http://arxiv.org/abs/2402.06900v2

cs.CL, cs.AI

GenTranslate: Large Language Models are Generative Multilingual Speech and Machine Translators

Yuchen Hu, Chen Chen, Chao-Han Huck Yang, Ruizhe Li, Dong Zhang, Zhehuai Chen, Eng Siong Chng

Recent advances in large language models (LLMs) have stepped forward the development of multilingual speech and machine translation by its reduced representation errors and incorporated external knowledge. However, both translation tasks typically utilize beam search decoding and top-1 hypothesis selection for inference. These techniques struggle to fully exploit the rich information in the diverse N-best hypotheses, making them less optimal for translation tasks that require a single, high-quality output sequence. In this paper, we propose a new generative paradigm for translation tasks, namely "GenTranslate", which builds upon LLMs to generate better results from the diverse translation versions in N-best list. Leveraging the rich linguistic knowledge and strong reasoning abilities of LLMs, our new paradigm can integrate the rich information in N-best candidates to generate a higher-quality translation result. Furthermore, to support LLM finetuning, we build and release a HypoTranslate dataset that contains over 592K hypotheses-translation pairs in 11 languages. Experiments on various speech and machine translation benchmarks (e.g., FLEURS, CoVoST-2, WMT) demonstrate that our GenTranslate significantly outperforms the state-of-the-art model.

http://arxiv.org/abs/2402.06894v1

cs.CL, cs.AI, cs.LG, cs.SD, eess.AS

UrbanKGent: A Unified Large Language Model Agent Framework for Urban Knowledge Graph Construction

Yansong Ning, Hao Liu

Urban knowledge graph has recently worked as an emerging building block to distill critical knowledge from multi-sourced urban data for diverse urban application scenarios. Despite its promising benefits, urban knowledge graph construction (UrbanKGC) still heavily relies on manual effort, hindering its potential advancement. This paper presents UrbanKGent, a unified large language model agent framework, for urban knowledge graph construction. Specifically, we first construct the knowledgeable instruction set for UrbanKGC tasks (such as relational triplet extraction and knowledge graph completion) via heterogeneity-aware and geospatial-infused instruction generation. Moreover, we propose a tool-augmented iterative trajectory refinement module to enhance and refine the trajectories distilled from GPT-4. Through hybrid instruction fine-tuning with augmented trajectories on Llama-2-13B, we obtain the UrbanKGC agent, UrbanKGent-13B. We perform a comprehensive evaluation on two real-world datasets using both human and GPT-4 self-evaluation. The experimental results demonstrate that UrbanKGent-13B not only can significantly outperform 21 baselines in UrbanKGC tasks, but also surpass the state-of-the-art LLM, GPT-4, by more than 10\% with approximately 20 times lower cost. We deploy UrbanKGent-13B to provide online services, which can construct an UrbanKG with thousands of times richer relationships using only one-fifth of the data compared with the existing benchmark. Our data, code, and opensource UrbanKGC agent are available at https://github.com/usail-hkust/UrbanKGent.

http://arxiv.org/abs/2402.06861v1

cs.AI

ChemLLM: A Chemical Large Language Model

Di Zhang, Wei Liu, Qian Tan, Jingdan Chen, Hang Yan, Yuliang Yan, Jiatong Li, Weiran Huang, Xiangyu Yue, Wanli Ouyang, Dongzhan Zhou, Shufei Zhang, Mao Su, Han-Sen Zhong, Yuqiang Li

Large language models (LLMs) have made impressive progress in chemistry applications. However, the community lacks an LLM specifically designed for chemistry. The main challenges are two-fold: firstly, most chemical data and scientific knowledge are stored in structured databases, which limits the model's ability to sustain coherent dialogue when used directly. Secondly, there is an absence of objective and fair benchmark that encompass most chemistry tasks. Here, we introduce ChemLLM, a comprehensive framework that features the first LLM dedicated to chemistry. It also includes ChemData, a dataset specifically designed for instruction tuning, and ChemBench, a robust benchmark covering nine essential chemistry tasks. ChemLLM is adept at performing various tasks across chemical disciplines with fluid dialogue interaction. Notably, ChemLLM achieves results comparable to GPT-4 on the core chemical tasks and demonstrates competitive performance with LLMs of similar size in general scenarios. ChemLLM paves a new path for exploration in chemical studies, and our method of incorporating structured chemical knowledge into dialogue systems sets a new standard for developing LLMs in various scientific fields. Codes, Datasets, and Model weights are publicly accessible at https://hf.co/AI4Chem

http://arxiv.org/abs/2402.06852v2

cs.AI, cs.CL

Forecasting Events in Soccer Matches Through Language

Tiago Mendes-Neves, Luís Meireles, João Mendes-Moreira

This paper introduces an approach to predicting the next event in a soccer match, a challenge bearing remarkable similarities to the problem faced by Large Language Models (LLMs). Unlike other methods that severely limit event dynamics in soccer, often abstracting from many variables or relying on a mix of sequential models, our research proposes a novel technique inspired by the methodologies used in LLMs. These models predict a complete chain of variables that compose an event, significantly simplifying the construction of Large Event Models (LEMs) for soccer. Utilizing deep learning on the publicly available WyScout dataset, the proposed approach notably surpasses the performance of previous LEM proposals in critical areas, such as the prediction accuracy of the next event type. This paper highlights the utility of LEMs in various applications, including match prediction and analytics. Moreover, we show that LEMs provide a simulation backbone for users to build many analytics pipelines, an approach opposite to the current specialized single-purpose models. LEMs represent a pivotal advancement in soccer analytics, establishing a foundational framework for multifaceted analytics pipelines through a singular machine-learning model.

http://arxiv.org/abs/2402.06820v2

cs.LG, cs.CL

The Unreasonable Effectiveness of Eccentric Automatic Prompts

Rick Battle, Teja Gollapudi

Large Language Models (LLMs) have demonstrated remarkable problem-solving and basic mathematics abilities. However, their efficacy is highly contingent on the formulation of the prompt. This study endeavors to quantify the influence of incorporating "positive thinking" into the system message of the prompt, then compare that to systematic prompt optimization. We assess the performance of 60 combinations of system message snippets, tested with and without Chain of Thought prompting, across three models with parameters ranging from 7 to 70 billion on the GSM8K dataset. Our findings reveal that results do not universally generalize across models. In most instances, the inclusion of "positive thinking" prompts positively affected model performance. Notably, however, Llama2-70B exhibited an exception when not utilizing Chain of Thought, as the optimal system message was found to be none at all. Given the combinatorial complexity, and thus computation time, of experimenting with hand-tuning prompts for large black-box models, we then compared the performance of the best "positive thinking" prompt against the output of systematic prompt optimization. We show that employing an automated prompt optimizer emerges as the most effective method for enhancing performance, even when working with smaller open-source models. Additionally, our findings reveal that the highest-scoring, automatically-optimized prompt exhibits a degree of peculiarity far beyond expectations.

http://arxiv.org/abs/2402.10949v2

cs.CL, cs.AI, cs.LG

Reasoning Grasping via Multimodal Large Language Model

Shiyu Jin, Jinxuan Xu, Yutian Lei, Liangjun Zhang

Despite significant progress in robotic systems for operation within human-centric environments, existing models still heavily rely on explicit human commands to identify and manipulate specific objects. This limits their effectiveness in environments where understanding and acting on implicit human intentions are crucial. In this study, we introduce a novel task: reasoning grasping, where robots need to generate grasp poses based on indirect verbal instructions or intentions. To accomplish this, we propose an end-to-end reasoning grasping model that integrates a multi-modal Large Language Model (LLM) with a vision-based robotic grasping framework. In addition, we present the first reasoning grasping benchmark dataset generated from the GraspNet-1 billion, incorporating implicit instructions for object-level and part-level grasping, and this dataset will soon be available for public access. Our results show that directly integrating CLIP or LLaVA with the grasp detection model performs poorly on the challenging reasoning grasping tasks, while our proposed model demonstrates significantly enhanced performance both in the reasoning grasping benchmark and real-world experiments.

http://arxiv.org/abs/2402.06798v2

cs.RO

GLaM: Fine-Tuning Large Language Models for Domain Knowledge Graph Alignment via Neighborhood Partitioning and Generative Subgraph Encoding

Stefan Dernbach, Khushbu Agarwal, Alejandro Zuniga, Michael Henry, Sutanay Choudhury

Integrating large language models (LLMs) with knowledge graphs derived from domain-specific data represents an important advancement towards more powerful and factual reasoning. As these models grow more capable, it is crucial to enable them to perform multi-step inferences over real-world knowledge graphs while minimizing hallucination. While large language models excel at conversation and text generation, their ability to reason over domain-specialized graphs of interconnected entities remains limited. For example, can we query a LLM to identify the optimal contact in a professional network for a specific goal, based on relationships and attributes in a private database? The answer is no--such capabilities lie beyond current methods. However, this question underscores a critical technical gap that must be addressed. Many high-value applications in areas such as science, security, and e-commerce rely on proprietary knowledge graphs encoding unique structures, relationships, and logical constraints. We introduce a fine-tuning framework for developing Graph-aligned LAnguage Models (GLaM) that transforms a knowledge graph into an alternate text representation with labeled question-answer pairs. We demonstrate that grounding the models in specific graph-based knowledge expands the models' capacity for structure-based reasoning. Our methodology leverages the large-language model's generative capabilities to create the dataset and proposes an efficient alternate to retrieval-augmented generation styled methods.

http://arxiv.org/abs/2402.06764v3

cs.AI

EntGPT: Linking Generative Large Language Models with Knowledge Bases

Yifan Ding, Amrit Poudel, Qingkai Zeng, Tim Weninger, Balaji Veeramani, Sanmitra Bhattacharya

The ability of Large Language Models (LLMs) to generate factually correct output remains relatively unexplored due to the lack of fact-checking and knowledge grounding during training and inference. In this work, we aim to address this challenge through the Entity Disambiguation (ED) task. We first consider prompt engineering, and design a three-step hard-prompting method to probe LLMs' ED performance without supervised fine-tuning (SFT). Overall, the prompting method improves the micro-F_1 score of the original vanilla models by a large margin, on some cases up to 36% and higher, and obtains comparable performance across 10 datasets when compared to existing methods with SFT. We further improve the knowledge grounding ability through instruction tuning (IT) with similar prompts and responses. The instruction-tuned model not only achieves higher micro-F1 score performance as compared to several baseline methods on supervised entity disambiguation tasks with an average micro-F_1 improvement of 2.1% over the existing baseline models, but also obtains higher accuracy on six Question Answering (QA) tasks in the zero-shot setting. Our methodologies apply to both open- and closed-source LLMs.

http://arxiv.org/abs/2402.06738v1

cs.CL

Feedback Loops With Language Models Drive In-Context Reward Hacking

Alexander Pan, Erik Jones, Meena Jagadeesan, Jacob Steinhardt

Language models influence the external world: they query APIs that read and write to web pages, generate content that shapes human behavior, and run system commands as autonomous agents. These interactions form feedback loops: LLM outputs affect the world, which in turn affect subsequent LLM outputs. In this work, we show that feedback loops can cause in-context reward hacking (ICRH), where the LLM at test-time optimizes a (potentially implicit) objective but creates negative side effects in the process. For example, consider an LLM agent deployed to increase Twitter engagement; the LLM may retrieve its previous tweets into the context window and make them more controversial, increasing engagement but also toxicity. We identify and study two processes that lead to ICRH: output-refinement and policy-refinement. For these processes, evaluations on static datasets are insufficient -- they miss the feedback effects and thus cannot capture the most harmful behavior. In response, we provide three recommendations for evaluation to capture more instances of ICRH. As AI development accelerates, the effects of feedback loops will proliferate, increasing the need to understand their role in shaping LLM behavior.

http://arxiv.org/abs/2402.06627v1

cs.LG, cs.AI, cs.CL

Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuning

Shivalika Singh, Freddie Vargus, Daniel Dsouza, Börje F. Karlsson, Abinaya Mahendiran, Wei-Yin Ko, Herumb Shandilya, Jay Patel, Deividas Mataciunas, Laura OMahony, Mike Zhang, Ramith Hettiarachchi, Joseph Wilson, Marina Machado, Luisa Souza Moura, Dominik Krzemiński, Hakimeh Fadaei, Irem Ergün, Ifeoma Okoh, Aisha Alaagib, Oshan Mudannayake, Zaid Alyafeai, Vu Minh Chien, Sebastian Ruder, Surya Guthikonda, Emad A. Alghamdi, Sebastian Gehrmann, Niklas Muennighoff, Max Bartolo, Julia Kreutzer, Ahmet Üstün, Marzieh Fadaee, Sara Hooker

Datasets are foundational to many breakthroughs in modern artificial intelligence. Many recent achievements in the space of natural language processing (NLP) can be attributed to the finetuning of pre-trained models on a diverse set of tasks that enables a large language model (LLM) to respond to instructions. Instruction fine-tuning (IFT) requires specifically constructed and annotated datasets. However, existing datasets are almost all in the English language. In this work, our primary goal is to bridge the language gap by building a human-curated instruction-following dataset spanning 65 languages. We worked with fluent speakers of languages from around the world to collect natural instances of instructions and completions. Furthermore, we create the most extensive multilingual collection to date, comprising 513 million instances through templating and translating existing datasets across 114 languages. In total, we contribute four key resources: we develop and open-source the Aya Annotation Platform, the Aya Dataset, the Aya Collection, and the Aya Evaluation Suite. The Aya initiative also serves as a valuable case study in participatory research, involving collaborators from 119 countries. We see this as a valuable framework for future research collaborations that aim to bridge gaps in resources.

http://arxiv.org/abs/2402.06619v1

cs.CL, cs.AI

TIC: Translate-Infer-Compile for accurate 'text to plan' using LLMs and logical intermediate representations

Sudhir Agarwal, Anu Sreepathy

We study the problem of generating plans for given natural language planning task requests. On one hand, LLMs excel at natural language processing but do not perform well on planning. On the other hand, classical planning tools excel at planning tasks but require input in a structured language such as the Planning Domain Definition Language (PDDL). We leverage the strengths of both the techniques by using an LLM for generating the PDDL representation (task PDDL) of planning task requests followed by using a classical planner for computing a plan. Unlike previous approaches that use LLMs for generating task PDDLs directly, our approach comprises of (a) translate: using an LLM only for generating a logically interpretable intermediate representation of natural language task descriptions, (b) infer: deriving additional logically dependent information from the intermediate representation using a logic reasoner (currently, Answer Set Programming solver), and (c) compile: generating the target task PDDL from the base and inferred information. We observe that using an LLM to only output the intermediate representation significantly reduces LLM errors. Consequently, TIC approach achieves, for at least one LLM, high accuracy on task PDDL generation for all seven domains of our evaluation dataset.

http://arxiv.org/abs/2402.06608v1

cs.CL, cs.AI

CigaR: Cost-efficient Program Repair with LLMs

Dávid Hidvégi, Khashayar Etemadi, Sofia Bobadilla, Martin Monperrus

Large language models (LLM) have proven to be effective at automated program repair (APR). However, using LLMs can be costly, with companies invoicing users by the number of tokens. In this paper, we propose CigaR, the first LLM-based APR tool that focuses on minimizing the repair cost. CigaR works in two major steps: generating a first plausible patch and multiplying plausible patches. CigaR optimizes the prompts and the prompt setting to maximize the information given to LLMs using the smallest possible number of tokens. Our experiments on 429 bugs from the widely used Defects4J and HumanEval-Java datasets shows that CigaR reduces the token cost by 73%. On average, CigaR spends 127k tokens per bug while the baseline uses 467k tokens per bug. On the subset of bugs that are fixed by both, CigaR spends 20k per bug while the baseline uses 608k tokens, a cost saving of 96%. Our extensive experiments show that CigaR is a cost-effective LLM-based program repair tool that uses a low number of tokens to automatically generate patches.

http://arxiv.org/abs/2402.06598v2

cs.SE

Calibrating Long-form Generations from Large Language Models

Yukun Huang, Yixin Liu, Raghuveer Thirukovalluru, Arman Cohan, Bhuwan Dhingra

To enhance Large Language Models' (LLMs) reliability, calibration is essential -- the model's assessed confidence scores should align with the actual likelihood of its responses being correct. However, current confidence elicitation methods and calibration metrics typically rely on a binary true/false assessment of response correctness. This approach does not apply to long-form generation, where an answer can be partially correct. Addressing this gap, we introduce a unified calibration framework, in which both the correctness of the LLMs' responses and their associated confidence levels are treated as distributions across a range of scores. Within this framework, we develop three metrics to precisely evaluate LLM calibration and further propose two confidence elicitation methods based on self-consistency and self-evaluation. Our experiments, which include long-form QA and summarization tasks, demonstrate that larger models don't necessarily guarantee better calibration, that calibration performance is found to be metric-dependent, and that self-consistency methods excel in factoid datasets. We also find that calibration can be enhanced through techniques such as fine-tuning, integrating relevant source documents, scaling the temperature, and combining self-consistency with self-evaluation. Lastly, we showcase a practical application of our system: selecting and cascading open-source models and ChatGPT to optimize correctness given a limited API budget. This research not only challenges existing notions of LLM calibration but also offers practical methodologies for improving trustworthiness in long-form generation.

http://arxiv.org/abs/2402.06544v1

cs.CL, cs.AI, cs.LG

StruQ: Defending Against Prompt Injection with Structured Queries

Sizhe Chen, Julien Piet, Chawin Sitawarin, David Wagner

Recent advances in Large Language Models (LLMs) enable exciting LLM-integrated applications, which perform text-based tasks by utilizing their advanced language understanding capabilities. However, as LLMs have improved, so have the attacks against them. Prompt injection attacks are an important threat: they trick the model to deviate from the original application's instructions and instead follow user directives. These attacks rely on the LLM's ability to follow instructions and inability to separate the prompts and user data. We introduce structured queries, a general approach to tackle this problem. Structured queries separate prompts and data into two channels. We implement a system that supports structured queries. This system is made of (1) a secure front-end that formats a prompt and user data into a special format, and (2) a specially trained LLM that can produce high-quality outputs from these inputs. The LLM is trained using a novel fine-tuning strategy: we convert a base (non-instruction-tuned) LLM to a structured instruction-tuned model that will only follow instructions in the prompt portion of a query. To do so, we augment standard instruction tuning datasets with examples that also include instructions in the data portion of the query, and fine-tune the model to ignore these. Our system significantly improves resistance to prompt injection attacks, with little or no impact on utility. Our code is released at https://github.com/Sizhe-Chen/PromptInjectionDefense.

http://arxiv.org/abs/2402.06363v1

cs.CR

RareBench: Can LLMs Serve as Rare Diseases Specialists?

Xuanzhong Chen, Xiaohao Mao, Qihan Guo, Lun Wang, Shuyang Zhang, Ting Chen

Generalist Large Language Models (LLMs), such as GPT-4, have shown considerable promise in various domains, including medical diagnosis. Rare diseases, affecting approximately 300 million people worldwide, often have unsatisfactory clinical diagnosis rates primarily due to a lack of experienced physicians and the complexity of differentiating among many rare diseases. In this context, recent news such as "ChatGPT correctly diagnosed a 4-year-old's rare disease after 17 doctors failed" underscore LLMs' potential, yet underexplored, role in clinically diagnosing rare diseases. To bridge this research gap, we introduce RareBench, a pioneering benchmark designed to systematically evaluate the capabilities of LLMs on 4 critical dimensions within the realm of rare diseases. Meanwhile, we have compiled the largest open-source dataset on rare disease patients, establishing a benchmark for future studies in this domain. To facilitate differential diagnosis of rare diseases, we develop a dynamic few-shot prompt methodology, leveraging a comprehensive rare disease knowledge graph synthesized from multiple knowledge bases, significantly enhancing LLMs' diagnostic performance. Moreover, we present an exhaustive comparative study of GPT-4's diagnostic capabilities against those of specialist physicians. Our experimental findings underscore the promising potential of integrating LLMs into the clinical diagnostic process for rare diseases. This paves the way for exciting possibilities in future advancements in this field.

http://arxiv.org/abs/2402.06341v1

cs.CL

ExaRanker-Open: Synthetic Explanation for IR using Open-Source LLMs

Fernando Ferraretto, Thiago Laitz, Roberto Lotufo, Rodrigo Nogueira

ExaRanker recently introduced an approach to training information retrieval (IR) models, incorporating natural language explanations as additional labels. The method addresses the challenge of limited labeled examples, leading to improvements in the effectiveness of IR models. However, the initial results were based on proprietary language models such as GPT-3.5, which posed constraints on dataset size due to its cost and data privacy. In this paper, we introduce ExaRanker-Open, where we adapt and explore the use of open-source language models to generate explanations. The method has been tested using different LLMs and datasets sizes to better comprehend the effective contribution of data augmentation. Our findings reveal that incorporating explanations consistently enhances neural rankers, with benefits escalating as the LLM size increases. Notably, the data augmentation method proves advantageous even with large datasets, as evidenced by ExaRanker surpassing the target baseline by 0.6 nDCG@10 points in our study. To encourage further advancements by the research community, we have open-sourced both the code and datasets at https://github.com/unicamp-dl/ExaRanker.

http://arxiv.org/abs/2402.06334v1

cs.IR, cs.AI, cs.CL

Understanding the Role of Cross-Entropy Loss in Fairly Evaluating Large Language Model-based Recommendation

Cong Xu, Zhangchi Zhu, Jun Wang, Jianyong Wang, Wei Zhang

Large language models (LLMs) have gained much attention in the recommendation community; some studies have observed that LLMs, fine-tuned by the cross-entropy loss with a full softmax, could achieve state-of-the-art performance already. However, these claims are drawn from unobjective and unfair comparisons. In view of the substantial quantity of items in reality, conventional recommenders typically adopt a pointwise/pairwise loss function instead for training. This substitute however causes severe performance degradation, leading to under-estimation of conventional methods and over-confidence in the ranking capability of LLMs. In this work, we theoretically justify the superiority of cross-entropy, and showcase that it can be adequately replaced by some elementary approximations with certain necessary modifications. The remarkable results across three public datasets corroborate that even in a practical sense, existing LLM-based methods are not as effective as claimed for next-item recommendation. We hope that these theoretical understandings in conjunction with the empirical results will facilitate an objective evaluation of LLM-based recommendation in the future.

http://arxiv.org/abs/2402.06216v2

cs.IR

The Generative AI Paradox on Evaluation: What It Can Solve, It May Not Evaluate

Juhyun Oh, Eunsu Kim, Inha Cha, Alice Oh

This paper explores the assumption that Large Language Models (LLMs) skilled in generation tasks are equally adept as evaluators. We assess the performance of three LLMs and one open-source LM in Question-Answering (QA) and evaluation tasks using the TriviaQA (Joshi et al., 2017) dataset. Results indicate a significant disparity, with LLMs exhibiting lower performance in evaluation tasks compared to generation tasks. Intriguingly, we discover instances of unfaithful evaluation where models accurately evaluate answers in areas where they lack competence, underscoring the need to examine the faithfulness and trustworthiness of LLMs as evaluators. This study contributes to the understanding of "the Generative AI Paradox" (West et al., 2023), highlighting a need to explore the correlation between generative excellence and evaluation proficiency, and the necessity to scrutinize the faithfulness aspect in model evaluations.

http://arxiv.org/abs/2402.06204v1

cs.CL, cs.AI

Large Language Models: A Survey

Shervin Minaee, Tomas Mikolov, Narjes Nikzad, Meysam Chenaghlu, Richard Socher, Xavier Amatriain, Jianfeng Gao

Large Language Models (LLMs) have drawn a lot of attention due to their strong performance on a wide range of natural language tasks, since the release of ChatGPT in November 2022. LLMs' ability of general-purpose language understanding and generation is acquired by training billions of model's parameters on massive amounts of text data, as predicted by scaling laws \cite{kaplan2020scaling,hoffmann2022training}. The research area of LLMs, while very recent, is evolving rapidly in many different ways. In this paper, we review some of the most prominent LLMs, including three popular LLM families (GPT, LLaMA, PaLM), and discuss their characteristics, contributions and limitations. We also give an overview of techniques developed to build, and augment LLMs. We then survey popular datasets prepared for LLM training, fine-tuning, and evaluation, review widely used LLM evaluation metrics, and compare the performance of several popular LLMs on a set of representative benchmarks. Finally, we conclude the paper by discussing open challenges and future research directions.

http://arxiv.org/abs/2402.06196v2

cs.CL, cs.AI

CultureLLM: Incorporating Cultural Differences into Large Language Models

Cheng Li, Mengzhou Chen, Jindong Wang, Sunayana Sitaram, Xing Xie

Large language models (LLMs) are reported to be partial to certain cultures owing to the training data dominance from the English corpora. Since multilingual cultural data are often expensive to collect, existing efforts handle this by prompt engineering or culture-specific pre-training. However, they might overlook the knowledge deficiency of low-resource culture and require extensive computing resources. In this paper, we propose CultureLLM, a cost-effective solution to incorporate cultural differences into LLMs. CultureLLM adopts World Value Survey (WVS) as seed data and generates semantically equivalent training data via the proposed semantic data augmentation. Using only 50 seed samples from WVS with augmented data, we fine-tune culture-specific LLMs and one unified model (CultureLLM-One) for 9 cultures covering rich and low-resource languages. Extensive experiments on 60 culture-related datasets demonstrate that CultureLLM significantly outperforms various counterparts such as GPT-3.5 (by 8.1%) and Gemini Pro (by 9.5%) with comparable performance to GPT-4 or even better. Our human study shows that the generated samples are semantically equivalent to the original samples, providing an effective solution for LLMs augmentation.

http://arxiv.org/abs/2402.10946v1

cs.CL, cs.AI, cs.LG

ScreenAgent: A Vision Language Model-driven Computer Control Agent

Runliang Niu, Jindong Li, Shiqi Wang, Yali Fu, Xiyu Hu, Xueyuan Leng, He Kong, Yi Chang, Qi Wang

Existing Large Language Models (LLM) can invoke a variety of tools and APIs to complete complex tasks. The computer, as the most powerful and universal tool, could potentially be controlled directly by a trained LLM agent. Powered by the computer, we can hopefully build a more generalized agent to assist humans in various daily digital works. In this paper, we construct an environment for a Vision Language Model (VLM) agent to interact with a real computer screen. Within this environment, the agent can observe screenshots and manipulate the Graphics User Interface (GUI) by outputting mouse and keyboard actions. We also design an automated control pipeline that includes planning, acting, and reflecting phases, guiding the agent to continuously interact with the environment and complete multi-step tasks. Additionally, we construct the ScreenAgent Dataset, which collects screenshots and action sequences when completing a variety of daily computer tasks. Finally, we trained a model, ScreenAgent, which achieved computer control capabilities comparable to GPT-4V and demonstrated more precise UI positioning capabilities. Our attempts could inspire further research on building a generalist LLM agent. The code is available at \url{https://github.com/niuzaisheng/ScreenAgent}.

http://arxiv.org/abs/2402.07945v1

cs.HC, cs.AI, cs.CV

Learn To be Efficient: Build Structured Sparsity in Large Language Models

Haizhong Zheng, Xiaoyan Bai, Beidi Chen, Fan Lai, Atul Prakash

Large Language Models (LLMs) have achieved remarkable success with their billion-level parameters, yet they incur high inference overheads. The emergence of activation sparsity in LLMs provides a natural approach to reduce this cost by involving only parts of the parameters for inference. Existing methods only focus on utilizing this naturally formed activation sparsity, overlooking the potential for further amplifying this inherent sparsity. In this paper, we hypothesize that LLMs can learn to be efficient by achieving more structured activation sparsity. To achieve this, we introduce a novel algorithm, Learn-To-be-Efficient (LTE), designed to train efficiency-aware LLMs to learn to activate fewer neurons and achieve a better trade-off between sparsity and performance. Furthermore, unlike SOTA MoEfication methods, which mainly focus on ReLU-based models, LTE can also be applied to LLMs like GPT and LLaMA with soft activation functions. We evaluate LTE on four models and eleven datasets. The experiments show that LTE achieves a better trade-off between sparsity and task performance. For instance, LTE with LLaMA provides a 1.83x-2.59x FLOPs speed-up on language generation tasks, outperforming the state-of-the-art methods.

http://arxiv.org/abs/2402.06126v2

cs.CL, cs.AI, cs.LG

Exploring Group and Symmetry Principles in Large Language Models

Shima Imani, Hamid Palangi

Large Language Models (LLMs) have demonstrated impressive performance across a wide range of applications; however, assessing their reasoning capabilities remains a significant challenge. In this paper, we introduce a framework grounded in group and symmetry principles, which have played a crucial role in fields such as physics and mathematics, and offer another way to evaluate their capabilities. While the proposed framework is general, to showcase the benefits of employing these properties, we focus on arithmetic reasoning and investigate the performance of these models on four group properties: closure, identity, inverse, and associativity. Our findings reveal that LLMs studied in this work struggle to preserve group properties across different test regimes. In the closure test, we observe biases towards specific outputs and an abrupt degradation in their performance from 100% to 0% after a specific sequence length. They also perform poorly in the identity test, which represents adding irrelevant information in the context, and show sensitivity when subjected to inverse test, which examines the robustness of the model with respect to negation. In addition, we demonstrate that breaking down problems into smaller steps helps LLMs in the associativity test that we have conducted. To support these tests we have developed a synthetic dataset which will be released.

http://arxiv.org/abs/2402.06120v1

cs.CL

SPHINX-X: Scaling Data and Parameters for a Family of Multi-modal Large Language Models

Peng Gao, Renrui Zhang, Chris Liu, Longtian Qiu, Siyuan Huang, Weifeng Lin, Shitian Zhao, Shijie Geng, Ziyi Lin, Peng Jin, Kaipeng Zhang, Wenqi Shao, Chao Xu, Conghui He, Junjun He, Hao Shao, Pan Lu, Hongsheng Li, Yu Qiao

We propose SPHINX-X, an extensive Multimodality Large Language Model (MLLM) series developed upon SPHINX. To improve the architecture and training efficiency, we modify the SPHINX framework by removing redundant visual encoders, bypassing fully-padded sub-images with skip tokens, and simplifying multi-stage training into a one-stage all-in-one paradigm. To fully unleash the potential of MLLMs, we assemble a comprehensive multi-domain and multimodal dataset covering publicly available resources in language, vision, and vision-language tasks. We further enrich this collection with our curated OCR intensive and Set-of-Mark datasets, extending the diversity and generality. By training over different base LLMs including TinyLlama1.1B, InternLM2-7B, LLaMA2-13B, and Mixtral8x7B, we obtain a spectrum of MLLMs that vary in parameter size and multilingual capabilities. Comprehensive benchmarking reveals a strong correlation between the multi-modal performance with the data and parameter scales. Code and models are released at https://github.com/Alpha-VLLM/LLaMA2-Accessory

http://arxiv.org/abs/2402.05935v1

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

Driving Everywhere with Large Language Model Policy Adaptation

Boyi Li, Yue Wang, Jiageng Mao, Boris Ivanovic, Sushant Veer, Karen Leung, Marco Pavone

Adapting driving behavior to new environments, customs, and laws is a long-standing problem in autonomous driving, precluding the widespread deployment of autonomous vehicles (AVs). In this paper, we present LLaDA, a simple yet powerful tool that enables human drivers and autonomous vehicles alike to drive everywhere by adapting their tasks and motion plans to traffic rules in new locations. LLaDA achieves this by leveraging the impressive zero-shot generalizability of large language models (LLMs) in interpreting the traffic rules in the local driver handbook. Through an extensive user study, we show that LLaDA's instructions are useful in disambiguating in-the-wild unexpected situations. We also demonstrate LLaDA's ability to adapt AV motion planning policies in real-world datasets; LLaDA outperforms baseline planning approaches on all our metrics. Please check our website for more details: https://boyiliee.github.io/llada.

http://arxiv.org/abs/2402.05932v2

cs.RO, cs.AI, cs.CL

FACT-GPT: Fact-Checking Augmentation via Claim Matching with LLMs

Eun Cheol Choi, Emilio Ferrara

Our society is facing rampant misinformation harming public health and trust. To address the societal challenge, we introduce FACT-GPT, a system leveraging Large Language Models (LLMs) to automate the claim matching stage of fact-checking. FACT-GPT, trained on a synthetic dataset, identifies social media content that aligns with, contradicts, or is irrelevant to previously debunked claims. Our evaluation shows that our specialized LLMs can match the accuracy of larger models in identifying related claims, closely mirroring human judgment. This research provides an automated solution for efficient claim matching, demonstrates the potential of LLMs in supporting fact-checkers, and offers valuable resources for further research in the field.

http://arxiv.org/abs/2402.05904v1

cs.CL, cs.CY, cs.HC, cs.SI

Large Language Model Meets Graph Neural Network in Knowledge Distillation

Shengxiang Hu, Guobing Zou, Song Yang, Yanglan Gan, Bofeng Zhang, Yixin Chen

Despite recent community revelations about the advancements and potential applications of Large Language Models (LLMs) in understanding Text-Attributed Graph (TAG), the deployment of LLMs for production is hindered by its high computational and storage requirements, as well as long latencies during model inference. Simultaneously, although traditional Graph Neural Networks (GNNs) are light weight and adept at learning structural features of graphs, their ability to grasp the complex semantics in TAG is somewhat constrained for real applications. To address these limitations, we concentrate on the downstream task of node classification in TAG and propose a novel graph knowledge distillation framework, termed Linguistic Graph Knowledge Distillation (LinguGKD), using LLMs as teacher models and GNNs as student models for knowledge distillation. It involves TAG-oriented instruction tuning of LLM on designed tailored prompts, followed by propagating knowledge and aligning the hierarchically learned node features from the teacher LLM to the student GNN in latent space, employing a layer-adaptive contrastive learning strategy. Through extensive experiments on a variety of LLM and GNN models and multiple benchmark datasets, the proposed LinguGKD significantly boosts the student GNN's predictive accuracy and convergence rate, without the need of extra data or model parameters. Compared to teacher LLM, distilled GNN achieves superior inference speed equipped with much fewer computing and storage demands, when surpassing the teacher LLM's classification accuracy on some of benchmark datasets.

http://arxiv.org/abs/2402.05894v2

cs.AI, cs.LG, 68T30, 68R10, 68T05

EmojiCrypt: Prompt Encryption for Secure Communication with Large Language Models

Guo Lin, Wenyue Hua, Yongfeng Zhang

Cloud-based large language models (LLMs) such as ChatGPT have increasingly become integral to daily operations, serving as vital tools across various applications. While these models offer substantial benefits in terms of accessibility and functionality, they also introduce significant privacy concerns: the transmission and storage of user data in cloud infrastructures pose substantial risks of data breaches and unauthorized access to sensitive information; even if the transmission and storage of data is encrypted, the LLM service provider itself still knows the real contents of the data, preventing individuals or entities from confidently using such LLM services. To address these concerns, this paper proposes a simple yet effective mechanism EmojiCrypt to protect user privacy. It uses Emoji to encrypt the user inputs before sending them to LLM, effectively rendering them indecipherable to human or LLM's examination while retaining the original intent of the prompt, thus ensuring the model's performance remains unaffected. We conduct experiments on three tasks, personalized recommendation, sentiment analysis, and tabular data analysis. Experiment results reveal that EmojiCrypt can encrypt personal information within prompts in such a manner that not only prevents the discernment of sensitive data by humans or LLM itself, but also maintains or even improves the precision without further tuning, achieving comparable or even better task accuracy than directly prompting the LLM without prompt encryption. These results highlight the practicality of adopting encryption measures that safeguard user privacy without compromising the functional integrity and performance of LLMs. Code and dataset are available at https://github.com/agiresearch/EmojiCrypt.

http://arxiv.org/abs/2402.05868v2

cs.CL, cs.AI, cs.CR, cs.IR, cs.LG

Examining Gender and Racial Bias in Large Vision-Language Models Using a Novel Dataset of Parallel Images

Kathleen C. Fraser, Svetlana Kiritchenko

Following on recent advances in large language models (LLMs) and subsequent chat models, a new wave of large vision-language models (LVLMs) has emerged. Such models can incorporate images as input in addition to text, and perform tasks such as visual question answering, image captioning, story generation, etc. Here, we examine potential gender and racial biases in such systems, based on the perceived characteristics of the people in the input images. To accomplish this, we present a new dataset PAIRS (PArallel Images for eveRyday Scenarios). The PAIRS dataset contains sets of AI-generated images of people, such that the images are highly similar in terms of background and visual content, but differ along the dimensions of gender (man, woman) and race (Black, white). By querying the LVLMs with such images, we observe significant differences in the responses according to the perceived gender or race of the person depicted.

http://arxiv.org/abs/2402.05779v1

cs.CY, cs.CL, cs.CV

Editable Scene Simulation for Autonomous Driving via Collaborative LLM-Agents

Yuxi Wei, Zi Wang, Yifan Lu, Chenxin Xu, Changxing Liu, Hao Zhao, Siheng Chen, Yanfeng Wang

Scene simulation in autonomous driving has gained significant attention because of its huge potential for generating customized data. However, existing editable scene simulation approaches face limitations in terms of user interaction efficiency, multi-camera photo-realistic rendering and external digital assets integration. To address these challenges, this paper introduces ChatSim, the first system that enables editable photo-realistic 3D driving scene simulations via natural language commands with external digital assets. To enable editing with high command flexibility,~ChatSim leverages a large language model (LLM) agent collaboration framework. To generate photo-realistic outcomes, ChatSim employs a novel multi-camera neural radiance field method. Furthermore, to unleash the potential of extensive high-quality digital assets, ChatSim employs a novel multi-camera lighting estimation method to achieve scene-consistent assets' rendering. Our experiments on Waymo Open Dataset demonstrate that ChatSim can handle complex language commands and generate corresponding photo-realistic scene videos.

http://arxiv.org/abs/2402.05746v2

cs.CV

Comprehensive Assessment of Jailbreak Attacks Against LLMs

Junjie Chu, Yugeng Liu, Ziqing Yang, Xinyue Shen, Michael Backes, Yang Zhang

Misuse of the Large Language Models (LLMs) has raised widespread concern. To address this issue, safeguards have been taken to ensure that LLMs align with social ethics. However, recent findings have revealed an unsettling vulnerability bypassing the safeguards of LLMs, known as jailbreak attacks. By applying techniques, such as employing role-playing scenarios, adversarial examples, or subtle subversion of safety objectives as a prompt, LLMs can produce an inappropriate or even harmful response. While researchers have studied several categories of jailbreak attacks, they have done so in isolation. To fill this gap, we present the first large-scale measurement of various jailbreak attack methods. We concentrate on 13 cutting-edge jailbreak methods from four categories, 160 questions from 16 violation categories, and six popular LLMs. Our extensive experimental results demonstrate that the optimized jailbreak prompts consistently achieve the highest attack success rates, as well as exhibit robustness across different LLMs. Some jailbreak prompt datasets, available from the Internet, can also achieve high attack success rates on many LLMs, such as ChatGLM3, GPT-3.5, and PaLM2. Despite the claims from many organizations regarding the coverage of violation categories in their policies, the attack success rates from these categories remain high, indicating the challenges of effectively aligning LLM policies and the ability to counter jailbreak attacks. We also discuss the trade-off between the attack performance and efficiency, as well as show that the transferability of the jailbreak prompts is still viable, becoming an option for black-box models. Overall, our research highlights the necessity of evaluating different jailbreak methods. We hope our study can provide insights for future research on jailbreak attacks and serve as a benchmark tool for evaluating them for practitioners.

http://arxiv.org/abs/2402.05668v1

cs.CR, cs.AI, cs.CL, cs.LG

GPT-4 Generated Narratives of Life Events using a Structured Narrative Prompt: A Validation Study

Christopher J. Lynch, Erik Jensen, Madison H. Munro, Virginia Zamponi, Joseph Martinez, Kevin O'Brien, Brandon Feldhaus, Katherine Smith, Ann Marie Reinhold, Ross Gore

Large Language Models (LLMs) play a pivotal role in generating vast arrays of narratives, facilitating a systematic exploration of their effectiveness for communicating life events in narrative form. In this study, we employ a zero-shot structured narrative prompt to generate 24,000 narratives using OpenAI's GPT-4. From this dataset, we manually classify 2,880 narratives and evaluate their validity in conveying birth, death, hiring, and firing events. Remarkably, 87.43% of the narratives sufficiently convey the intention of the structured prompt. To automate the identification of valid and invalid narratives, we train and validate nine Machine Learning models on the classified datasets. Leveraging these models, we extend our analysis to predict the classifications of the remaining 21,120 narratives. All the ML models excelled at classifying valid narratives as valid, but experienced challenges at simultaneously classifying invalid narratives as invalid. Our findings not only advance the study of LLM capabilities, limitations, and validity but also offer practical insights for narrative generation and natural language processing applications.

http://arxiv.org/abs/2402.05435v1

cs.CL, cs.AI, cs.LG, I.2.7; I.6.4

Zero-Shot Chain-of-Thought Reasoning Guided by Evolutionary Algorithms in Large Language Models

Feihu Jin, Yifan Liu, Ying Tan

Large Language Models (LLMs) have demonstrated remarkable performance across diverse tasks and exhibited impressive reasoning abilities by applying zero-shot Chain-of-Thought (CoT) prompting. However, due to the evolving nature of sentence prefixes during the pre-training phase, existing zero-shot CoT prompting methods that employ identical CoT prompting across all task instances may not be optimal. In this paper, we introduce a novel zero-shot prompting method that leverages evolutionary algorithms to generate diverse promptings for LLMs dynamically. Our approach involves initializing two CoT promptings, performing evolutionary operations based on LLMs to create a varied set, and utilizing the LLMs to select a suitable CoT prompting for a given problem. Additionally, a rewriting operation, guided by the selected CoT prompting, enhances the understanding of the LLMs about the problem. Extensive experiments conducted across ten reasoning datasets demonstrate the superior performance of our proposed method compared to current zero-shot CoT prompting methods on GPT-3.5-turbo and GPT-4. Moreover, in-depth analytical experiments underscore the adaptability and effectiveness of our method in various reasoning tasks.

http://arxiv.org/abs/2402.05376v1

cs.CL

CIC: A framework for Culturally-aware Image Captioning

Youngsik Yun, Jihie Kim

Image Captioning generates descriptive sentences from images using Vision-Language Pre-trained models (VLPs) such as BLIP, which has improved greatly. However, current methods lack the generation of detailed descriptive captions for the cultural elements depicted in the images, such as the traditional clothing worn by people from Asian cultural groups. In this paper, we propose a new framework, \textbf{Culturally-aware Image Captioning (CIC)}, that generates captions and describes cultural elements extracted from cultural visual elements in images representing cultures. Inspired by methods combining visual modality and Large Language Models (LLMs) through appropriate prompts, our framework (1) generates questions based on cultural categories from images, (2) extracts cultural visual elements from Visual Question Answering (VQA) using generated questions, and (3) generates culturally-aware captions using LLMs with the prompts. Our human evaluation conducted on 45 participants from 4 different cultural groups with a high understanding of the corresponding culture shows that our proposed framework generates more culturally descriptive captions when compared to the image captioning baseline based on VLPs. Our code and dataset will be made publicly available upon acceptance.

http://arxiv.org/abs/2402.05374v2

cs.CV, cs.AI, cs.CL

Scaling Up LLM Reviews for Google Ads Content Moderation

Wei Qiao, Tushar Dogra, Otilia Stretcu, Yu-Han Lyu, Tiantian Fang, Dongjin Kwon, Chun-Ta Lu, Enming Luo, Yuan Wang, Chih-Chun Chia, Ariel Fuxman, Fangzhou Wang, Ranjay Krishna, Mehmet Tek

Large language models (LLMs) are powerful tools for content moderation, but their inference costs and latency make them prohibitive for casual use on large datasets, such as the Google Ads repository. This study proposes a method for scaling up LLM reviews for content moderation in Google Ads. First, we use heuristics to select candidates via filtering and duplicate removal, and create clusters of ads for which we select one representative ad per cluster. We then use LLMs to review only the representative ads. Finally, we propagate the LLM decisions for the representative ads back to their clusters. This method reduces the number of reviews by more than 3 orders of magnitude while achieving a 2x recall compared to a baseline non-LLM model. The success of this approach is a strong function of the representations used in clustering and label propagation; we found that cross-modal similarity representations yield better results than uni-modal representations.

http://arxiv.org/abs/2402.14590v1

cs.IR, cs.CL, cs.LG

Are LLMs Ready for Real-World Materials Discovery?

Santiago Miret, N M Anoop Krishnan

Large Language Models (LLMs) create exciting possibilities for powerful language processing tools to accelerate research in materials science. While LLMs have great potential to accelerate materials understanding and discovery, they currently fall short in being practical materials science tools. In this position paper, we show relevant failure cases of LLMs in materials science that reveal current limitations of LLMs related to comprehending and reasoning over complex, interconnected materials science knowledge. Given those shortcomings, we outline a framework for developing Materials Science LLMs (MatSci-LLMs) that are grounded in materials science knowledge and hypothesis generation followed by hypothesis testing. The path to attaining performant MatSci-LLMs rests in large part on building high-quality, multi-modal datasets sourced from scientific literature where various information extraction challenges persist. As such, we describe key materials science information extraction challenges which need to be overcome in order to build large-scale, multi-modal datasets that capture valuable materials science knowledge. Finally, we outline a roadmap for applying future MatSci-LLMs for real-world materials discovery via: 1. Automated Knowledge Base Generation; 2. Automated In-Silico Material Design; and 3. MatSci-LLM Integrated Self-Driving Materials Laboratories.

http://arxiv.org/abs/2402.05200v1

cond-mat.mtrl-sci, cs.AI, cs.CL, cs.LG

Pedagogical Alignment of Large Language Models

Shashank Sonkar, Kangqi Ni, Sapana Chaudhary, Richard G. Baraniuk

In this paper, we introduce the novel concept of pedagogically aligned Large Language Models (LLMs) that signifies a transformative shift in the application of LLMs within educational contexts. Rather than providing direct responses to user queries, pedagogically-aligned LLMs function as scaffolding tools, breaking complex problems into manageable subproblems and guiding students towards the final answer through constructive feedback and hints. The objective is to equip learners with problem-solving strategies that deepen their understanding and internalization of the subject matter. Previous research in this field has primarily applied the supervised finetuning approach without framing the objective as an alignment problem, hence not employing reinforcement learning through human feedback (RLHF) methods. This study reinterprets the narrative by viewing the task through the lens of alignment and demonstrates how RLHF methods emerge naturally as a superior alternative for aligning LLM behaviour. Building on this perspective, we propose a novel approach for constructing a reward dataset specifically designed for the pedagogical alignment of LLMs. We apply three state-of-the-art RLHF algorithms and find that they outperform SFT significantly. Our qualitative analyses across model differences and hyperparameter sensitivity further validate the superiority of RLHF over SFT. Also, our study sheds light on the potential of online feedback for enhancing the performance of pedagogically-aligned LLMs, thus providing valuable insights for the advancement of these models in educational settings.

http://arxiv.org/abs/2402.05000v1

cs.CL

An Enhanced Prompt-Based LLM Reasoning Scheme via Knowledge Graph-Integrated Collaboration

Yihao Li, Ru Zhang, Jianyi Liu, Gongshen Liu

While Large Language Models (LLMs) demonstrate exceptional performance in a multitude of Natural Language Processing (NLP) tasks, they encounter challenges in practical applications, including issues with hallucinations, inadequate knowledge updating, and limited transparency in the reasoning process. To overcome these limitations, this study innovatively proposes a collaborative training-free reasoning scheme involving tight cooperation between Knowledge Graph (KG) and LLMs. This scheme first involves using LLMs to iteratively explore KG, selectively retrieving a task-relevant knowledge subgraph to support reasoning. The LLMs are then guided to further combine inherent implicit knowledge to reason on the subgraph while explicitly elucidating the reasoning process. Through such a cooperative approach, our scheme achieves more reliable knowledge-based reasoning and facilitates the tracing of the reasoning results. Experimental results show that our scheme significantly progressed across multiple datasets, notably achieving over a 10% improvement on the QALD10 dataset compared to the best baseline and the fine-tuned state-of-the-art (SOTA) work. Building on this success, this study hopes to offer a valuable reference for future research in the fusion of KG and LLMs, thereby enhancing LLMs' proficiency in solving complex issues.

http://arxiv.org/abs/2402.04978v1

cs.CL, cs.AI

Reconfidencing LLMs from the Grouping Loss Perspective

Lihu Chen, Alexandre Perez-Lebel, Fabian M. Suchanek, Gaël Varoquaux

Large Language Models (LLMs), including ChatGPT and LLaMA, are susceptible to generating hallucinated answers in a confident tone. While efforts to elicit and calibrate confidence scores have proven useful, recent findings show that controlling uncertainty must go beyond calibration: predicted scores may deviate significantly from the actual posterior probabilities due to the impact of grouping loss. In this work, we construct a new evaluation dataset derived from a knowledge base to assess confidence scores given to answers of Mistral and LLaMA. Experiments show that they tend to be overconfident. Further, we show that they are more overconfident on some answers than others, \emph{eg} depending on the nationality of the person in the query. In uncertainty-quantification theory, this is grouping loss. To address this, we propose a solution to reconfidence LLMs, canceling not only calibration but also grouping loss. The LLMs, after the reconfidencing process, indicate improved confidence alignment with the accuracy of their responses.

http://arxiv.org/abs/2402.04957v1

cs.CL

L4Q: Parameter Efficient Quantization-Aware Training on Large Language Models via LoRA-wise LSQ

Hyesung Jeon, Yulhwa Kim, Jae-joon Kim

Post-training quantization (PTQ) and quantization-aware training (QAT) methods are gaining popularity in mitigating the high memory and computational costs associated with Large Language Models (LLMs). In resource-constrained scenarios, PTQ, with its reduced training overhead, is often preferred over QAT, despite the latter's potential for higher accuracy. Meanwhile, parameter-efficient fine-tuning (PEFT) methods like low-rank adaptation (LoRA) have been introduced, and recent efforts have explored quantization-aware PEFT techniques. However, these approaches may lack generality due to their reliance on the pre-quantized model's configuration. Their effectiveness may be compromised by non-linearly quantized or mixed-precision weights, and the retraining of specific quantization parameters might impede optimal performance. To address these challenges, we propose L4Q, an algorithm for parameter-efficient quantization-aware training. L4Q leverages LoRA-wise learned quantization step size for LLMs, aiming to enhance generality. The simultaneous quantization-and-fine-tuning process of L4Q is applicable to high-precision models, yielding linearly quantized weights with superior accuracy. Our experiments, conducted on the LLaMA and LLaMA2 model families using an instructional dataset, showcase L4Q's capabilities in language comprehension and few-shot in-context learning, achieving sub-4-bit precision while maintaining comparable training times to applying PEFT on a quantized model.

http://arxiv.org/abs/2402.04902v2

cs.LG, cs.CL

Detecting Generated Native Ads in Conversational Search

Sebastian Schmidt, Ines Zelch, Janek Bevendorff, Benno Stein, Matthias Hagen, Martin Potthast

Conversational search engines such as YouChat and Microsoft Copilot use large language models (LLMs) to generate responses to queries. It is only a small step to also let the same technology insert ads within the generated responses - instead of separately placing ads next to a response. Inserted ads would be reminiscent of native advertising and product placement, both of which are very effective forms of subtle and manipulative advertising. Considering the high computational costs associated with LLMs, for which providers need to develop sustainable business models, users of conversational search engines may very well be confronted with generated native ads in the near future. In this paper, we thus take a first step to investigate whether LLMs can also be used as a countermeasure, i.e., to block generated native ads. We compile the Webis Generated Native Ads 2024 dataset of queries and generated responses with automatically inserted ads, and evaluate whether LLMs or fine-tuned sentence transformers can detect the ads. In our experiments, the investigated LLMs struggle with the task but sentence transformers achieve precision and recall values above 0.9.

http://arxiv.org/abs/2402.04889v2

cs.IR, cs.CL

Leveraging LLMs for Unsupervised Dense Retriever Ranking

Ekaterina Khramtsova, Shengyao Zhuang, Mahsa Baktashmotlagh, Guido Zuccon

This paper introduces a novel unsupervised technique that utilizes large language models (LLMs) to determine the most suitable dense retriever for a specific test(target) corpus. Selecting the appropriate dense retriever is vital for numerous IR applications that employ these retrievers, trained on public datasets, to encode or conduct searches within a new private target corpus. The effectiveness of a dense retriever can significantly diminish when applied to a target corpus that diverges in domain or task from the original training set. The problem becomes more pronounced in cases where the target corpus is unlabeled, e.g. in zero-shot scenarios, rendering direct evaluation of the model's effectiveness on the target corpus unattainable. Therefore, the unsupervised selection of an optimally pre-trained dense retriever, especially under conditions of domain shift, emerges as a critical challenge. Existing methodologies for ranking dense retrievers fall short in addressing these domain shift scenarios. To tackle this, our method capitalizes on LLMs to create pseudo-relevant queries, labels, and reference lists by analyzing a subset of documents from the target corpus. This allows for the ranking of dense retrievers based on their performance with these pseudo-relevant signals. Significantly, this strategy is the first to depend exclusively on the target corpus data, removing the necessity for training data and test labels. We assessed the effectiveness of our approach by compiling a comprehensive pool of cutting-edge dense retrievers and comparing our method against traditional dense retriever selection benchmarks. The findings reveal that our proposed solution surpasses the existing benchmarks in both the selection and ranking of dense retrievers.

http://arxiv.org/abs/2402.04853v1

cs.IR

Multi-Patch Prediction: Adapting LLMs for Time Series Representation Learning

Yuxuan Bian, Xuan Ju, Jiangtong Li, Zhijian Xu, Dawei Cheng, Qiang Xu

In this study, we present aLLM4TS, an innovative framework that adapts Large Language Models (LLMs) for time-series representation learning. Central to our approach is that we reconceive time-series forecasting as a self-supervised, multi-patch prediction task, which, compared to traditional contrastive learning or mask-and-reconstruction methods, captures temporal dynamics in patch representations more effectively. Our strategy encompasses two-stage training: (i). a causal continual pre-training phase on various time-series datasets, anchored on next patch prediction, effectively syncing LLM capabilities with the intricacies of time-series data; (ii). fine-tuning for multi-patch prediction in the targeted time-series context. A distinctive element of our framework is the patch-wise decoding layer, which departs from previous methods reliant on sequence-level decoding. Such a design directly transposes individual patches into temporal sequences, thereby significantly bolstering the model's proficiency in mastering temporal patch-based representations. aLLM4TS demonstrates superior performance in several downstream tasks, proving its effectiveness in deriving temporal representations with enhanced transferability and marking a pivotal advancement in the adaptation of LLMs for time-series analysis.

http://arxiv.org/abs/2402.04852v2

cs.LG

PaDeLLM-NER: Parallel Decoding in Large Language Models for Named Entity Recognition

Jinghui Lu, Ziwei Yang, Yanjie Wang, Xuejing Liu, Brian Mac Namee, Can Huang

In this study, we aim to reduce generation latency for Named Entity Recognition (NER) with Large Language Models (LLMs). The main cause of high latency in LLMs is the sequential decoding process, which autoregressively generates all labels and mentions for NER, significantly increase the sequence length. To this end, we introduce Parallel Decoding in LLM for NE} (PaDeLLM-NER), a approach that integrates seamlessly into existing generative model frameworks without necessitating additional modules or architectural modifications. PaDeLLM-NER allows for the simultaneous decoding of all mentions, thereby reducing generation latency. Experiments reveal that PaDeLLM-NER significantly increases inference speed that is 1.76 to 10.22 times faster than the autoregressive approach for both English and Chinese. Simultaneously it maintains the quality of predictions as evidenced by the performance that is on par with the state-of-the-art across various datasets.

http://arxiv.org/abs/2402.04838v4

cs.CL, cs.AI

Long Is More for Alignment: A Simple but Tough-to-Beat Baseline for Instruction Fine-Tuning

Hao Zhao, Maksym Andriushchenko, Francesco Croce, Nicolas Flammarion

There is a consensus that instruction fine-tuning of LLMs requires high-quality data, but what are they? LIMA (NeurIPS 2023) and AlpaGasus (ICLR 2024) are state-of-the-art methods for selecting such high-quality examples, either via manual curation or using GPT-3.5-Turbo as a quality scorer. We show that the extremely simple baseline of selecting the 1,000 instructions with longest responses from standard datasets can consistently outperform these sophisticated methods according to GPT-4 and PaLM-2 as judges, while remaining competitive on the OpenLLM benchmarks that test factual knowledge. We demonstrate this for several state-of-the-art LLMs (Llama-2-7B, Llama-2-13B, and Mistral-7B) and datasets (Alpaca-52k and Evol-Instruct-70k). In addition, a lightweight refinement of such long instructions can further improve the abilities of the fine-tuned LLMs, and allows us to obtain the 2nd highest-ranked Llama-2-7B-based model on AlpacaEval 2.0 while training on only 1,000 examples and no extra preference data. We also conduct a thorough analysis of our models to ensure that their enhanced performance is not simply due to GPT-4's preference for longer responses, thus ruling out any artificial improvement. In conclusion, our findings suggest that fine-tuning on the longest instructions should be the default baseline for any research on instruction fine-tuning.

http://arxiv.org/abs/2402.04833v1

cs.CL

Direct Language Model Alignment from Online AI Feedback

Shangmin Guo, Biao Zhang, Tianlin Liu, Tianqi Liu, Misha Khalman, Felipe Llinares, Alexandre Rame, Thomas Mesnard, Yao Zhao, Bilal Piot, Johan Ferret, Mathieu Blondel

Direct alignment from preferences (DAP) methods, such as DPO, have recently emerged as efficient alternatives to reinforcement learning from human feedback (RLHF), that do not require a separate reward model. However, the preference datasets used in DAP methods are usually collected ahead of training and never updated, thus the feedback is purely offline. Moreover, responses in these datasets are often sampled from a language model distinct from the one being aligned, and since the model evolves over training, the alignment phase is inevitably off-policy. In this study, we posit that online feedback is key and improves DAP methods. Our method, online AI feedback (OAIF), uses an LLM as annotator: on each training iteration, we sample two responses from the current model and prompt the LLM annotator to choose which one is preferred, thus providing online feedback. Despite its simplicity, we demonstrate via human evaluation in several tasks that OAIF outperforms both offline DAP and RLHF methods. We further show that the feedback leveraged in OAIF is easily controllable, via instruction prompts to the LLM annotator.

http://arxiv.org/abs/2402.04792v2

cs.AI, cs.CL, cs.HC

MLLM-as-a-Judge: Assessing Multimodal LLM-as-a-Judge with Vision-Language Benchmark

Dongping Chen, Ruoxi Chen, Shilin Zhang, Yinuo Liu, Yaochen Wang, Huichi Zhou, Qihui Zhang, Pan Zhou, Yao Wan, Lichao Sun

Multimodal Large Language Models (MLLMs) have gained significant attention recently, showing remarkable potential in artificial general intelligence. However, assessing the utility of MLLMs presents considerable challenges, primarily due to the absence multimodal benchmarks that align with human preferences. Inspired by LLM-as-a-Judge in LLMs, this paper introduces a novel benchmark, termed MLLM-as-a-Judge, to assess the ability of MLLMs in assisting judges including three distinct tasks: Scoring Evaluation, Pair Comparison, and Batch Ranking. Our study reveals that, while MLLMs demonstrate remarkable human-like discernment in Pair Comparisons, there is a significant divergence from human preferences in Scoring Evaluation and Batch Ranking tasks. Furthermore, MLLMs still face challenges in judgment, including diverse biases, hallucinatory responses, and inconsistencies, even for advanced models such as GPT-4V. These findings emphasize the pressing need for enhancements and further research efforts regarding MLLMs as fully reliable evaluators. Code and dataset are available at https://github.com/Dongping-Chen/MLLM-as-a-Judge.

http://arxiv.org/abs/2402.04788v1

cs.CL, cs.AI, cs.CV

Large Language Models As Faithful Explainers

Yu-Neng Chuang, Guanchu Wang, Chia-Yuan Chang, Ruixiang Tang, Fan Yang, Mengnan Du, Xuanting Cai, Xia Hu

Large Language Models (LLMs) have recently become proficient in addressing complex tasks by utilizing their rich internal knowledge and reasoning ability. Consequently, this complexity hinders traditional input-focused explanation algorithms for explaining the complex decision-making processes of LLMs. Recent advancements have thus emerged for self-explaining their predictions through a single feed-forward inference in a natural language format. However, natural language explanations are often criticized for lack of faithfulness since these explanations may not accurately reflect the decision-making behaviors of the LLMs. In this work, we introduce a generative explanation framework, xLLM, to improve the faithfulness of the explanations provided in natural language formats for LLMs. Specifically, we propose an evaluator to quantify the faithfulness of natural language explanation and enhance the faithfulness by an iterative optimization process of xLLM, with the goal of maximizing the faithfulness scores. Experiments conducted on three NLU datasets demonstrate that xLLM can significantly improve the faithfulness of generated explanations, which are in alignment with the behaviors of LLMs.

http://arxiv.org/abs/2402.04678v1

cs.CL, cs.AI, cs.LG

TransLLaMa: LLM-based Simultaneous Translation System

Roman Koshkin, Katsuhito Sudoh, Satoshi Nakamura

Decoder-only large language models (LLMs) have recently demonstrated impressive capabilities in text generation and reasoning. Nonetheless, they have limited applications in simultaneous machine translation (SiMT), currently dominated by encoder-decoder transformers. This study demonstrates that, after fine-tuning on a small dataset comprising causally aligned source and target sentence pairs, a pre-trained open-source LLM can control input segmentation directly by generating a special "wait" token. This obviates the need for a separate policy and enables the LLM to perform English-German and English-Russian SiMT tasks with BLEU scores that are comparable to those of specific state-of-the-art baselines. We also evaluated closed-source models such as GPT-4, which displayed encouraging results in performing the SiMT task without prior training (zero-shot), indicating a promising avenue for enhancing future SiMT systems.

http://arxiv.org/abs/2402.04636v1

cs.CL

SPARQL Generation: an analysis on fine-tuning OpenLLaMA for Question Answering over a Life Science Knowledge Graph

Julio C. Rangel, Tarcisio Mendes de Farias, Ana Claudia Sima, Norio Kobayashi

The recent success of Large Language Models (LLM) in a wide range of Natural Language Processing applications opens the path towards novel Question Answering Systems over Knowledge Graphs leveraging LLMs. However, one of the main obstacles preventing their implementation is the scarcity of training data for the task of translating questions into corresponding SPARQL queries, particularly in the case of domain-specific KGs. To overcome this challenge, in this study, we evaluate several strategies for fine-tuning the OpenLlama LLM for question answering over life science knowledge graphs. In particular, we propose an end-to-end data augmentation approach for extending a set of existing queries over a given knowledge graph towards a larger dataset of semantically enriched question-to-SPARQL query pairs, enabling fine-tuning even for datasets where these pairs are scarce. In this context, we also investigate the role of semantic "clues" in the queries, such as meaningful variable names and inline comments. Finally, we evaluate our approach over the real-world Bgee gene expression knowledge graph and we show that semantic clues can improve model performance by up to 33% compared to a baseline with random variable names and no comments included.

http://arxiv.org/abs/2402.04627v1

cs.AI, cs.CL, cs.DB, cs.IR

TinyLLM: Learning a Small Student from Multiple Large Language Models

Yijun Tian, Yikun Han, Xiusi Chen, Wei Wang, Nitesh V. Chawla

Transferring the reasoning capability from stronger large language models (LLMs) to smaller ones has been quite appealing, as smaller LLMs are more flexible to deploy with less expense. Among the existing solutions, knowledge distillation stands out due to its outstanding efficiency and generalization. However, existing methods suffer from several drawbacks, including limited knowledge diversity and the lack of rich contextual information. To solve the problems and facilitate the learning of compact language models, we propose TinyLLM, a new knowledge distillation paradigm to learn a small student LLM from multiple large teacher LLMs. In particular, we encourage the student LLM to not only generate the correct answers but also understand the rationales behind these answers. Given that different LLMs possess diverse reasoning skills, we guide the student model to assimilate knowledge from various teacher LLMs. We further introduce an in-context example generator and a teacher-forcing Chain-of-Thought strategy to ensure that the rationales are accurate and grounded in contextually appropriate scenarios. Extensive experiments on six datasets across two reasoning tasks demonstrate the superiority of our method. Results show that TinyLLM can outperform large teacher LLMs significantly, despite a considerably smaller model size.

http://arxiv.org/abs/2402.04616v2

cs.CL, cs.AI, cs.LG

Alirector: Alignment-Enhanced Chinese Grammatical Error Corrector

Haihui Yang, Xiaojun Quan

Chinese grammatical error correction (CGEC) faces serious overcorrection challenges when employing autoregressive generative models such as sequence-to-sequence (Seq2Seq) models and decoder-only large language models (LLMs). While previous methods aim to address overcorrection in Seq2Seq models, they are difficult to adapt to decoder-only LLMs. In this paper, we propose an alignment-enhanced corrector for the overcorrection problem that applies to both Seq2Seq models and decoder-only LLMs. Our method first trains a correction model to generate an initial correction of the source sentence. Then, we combine the source sentence with the initial correction and feed it through an alignment model for another round of correction, aiming to enforce the alignment model to focus on potential overcorrection. Moreover, to enhance the model's ability to identify nuances, we further explore the reverse alignment of the source sentence and the initial correction. Finally, we transfer the alignment knowledge from two alignment models to the correction model, instructing it on how to avoid overcorrection. Experimental results on three CGEC datasets demonstrate the effectiveness of our approach in alleviating overcorrection and improving overall performance.

http://arxiv.org/abs/2402.04601v1

cs.CL, cs.AI

UltraLink: An Open-Source Knowledge-Enhanced Multilingual Supervised Fine-tuning Dataset

Haoyu Wang, Shuo Wang, Yukun Yan, Xujia Wang, Zhiyu Yang, Yuzhuang Xu, Zhenghao Liu, Liner Yang, Ning Ding, Xu Han, Zhiyuan Liu, Maosong Sun

Open-source large language models (LLMs) have gained significant strength across diverse fields. Nevertheless, the majority of studies primarily concentrate on English, with only limited exploration into the realm of multilingual abilities. In this work, we therefore construct an open-source multilingual supervised fine-tuning dataset. Different from previous works that simply translate English instructions, we consider both the language-specific and language-agnostic abilities of LLMs. Firstly, we introduce a knowledge-grounded data augmentation approach to elicit more language-specific knowledge of LLMs, improving their ability to serve users from different countries. Moreover, we find modern LLMs possess strong cross-lingual transfer capabilities, thus repeatedly learning identical content in various languages is not necessary. Consequently, we can substantially prune the language-agnostic supervised fine-tuning (SFT) data without any performance degradation, making multilingual SFT more efficient. The resulting UltraLink dataset comprises approximately 1 million samples across five languages (i.e., En, Zh, Ru, Fr, Es), and the proposed data construction method can be easily extended to other languages. UltraLink-LM, which is trained on UltraLink, outperforms several representative baselines across many tasks.

http://arxiv.org/abs/2402.04588v2

cs.CL

SumRec: A Framework for Recommendation using Open-Domain Dialogue

Ryutaro Asahara, Masaki Takahashi, Chiho Iwahashi, Michimasa Inaba

Chat dialogues contain considerable useful information about a speaker's interests, preferences, and experiences.Thus, knowledge from open-domain chat dialogue can be used to personalize various systems and offer recommendations for advanced information.This study proposed a novel framework SumRec for recommending information from open-domain chat dialogue.The study also examined the framework using ChatRec, a newly constructed dataset for training and evaluation. To extract the speaker and item characteristics, the SumRec framework employs a large language model (LLM) to generate a summary of the speaker information from a dialogue and to recommend information about an item according to the type of user.The speaker and item information are then input into a score estimation model, generating a recommendation score.Experimental results show that the SumRec framework provides better recommendations than the baseline method of using dialogues and item descriptions in their original form. Our dataset and code is publicly available at https://github.com/Ryutaro-A/SumRec

http://arxiv.org/abs/2402.04523v1

cs.LG, cs.CL

LegalLens: Leveraging LLMs for Legal Violation Identification in Unstructured Text

Dor Bernsohn, Gil Semo, Yaron Vazana, Gila Hayat, Ben Hagag, Joel Niklaus, Rohit Saha, Kyryl Truskovskyi

In this study, we focus on two main tasks, the first for detecting legal violations within unstructured textual data, and the second for associating these violations with potentially affected individuals. We constructed two datasets using Large Language Models (LLMs) which were subsequently validated by domain expert annotators. Both tasks were designed specifically for the context of class-action cases. The experimental design incorporated fine-tuning models from the BERT family and open-source LLMs, and conducting few-shot experiments using closed-source LLMs. Our results, with an F1-score of 62.69\% (violation identification) and 81.02\% (associating victims), show that our datasets and setups can be used for both tasks. Finally, we publicly release the datasets and the code used for the experiments in order to advance further research in the area of legal natural language processing (NLP).

http://arxiv.org/abs/2402.04335v1

cs.CL, cs.AI, cs.LG

LESS: Selecting Influential Data for Targeted Instruction Tuning

Mengzhou Xia, Sadhika Malladi, Suchin Gururangan, Sanjeev Arora, Danqi Chen

Instruction tuning has unlocked powerful capabilities in large language models (LLMs), effectively using combined datasets to develop generalpurpose chatbots. However, real-world applications often require a specialized suite of skills (e.g., reasoning). The challenge lies in identifying the most relevant data from these extensive datasets to effectively develop specific capabilities, a setting we frame as targeted instruction tuning. We propose LESS, an optimizer-aware and practically efficient algorithm to effectively estimate data influences and perform Low-rank gradiEnt Similarity Search for instruction data selection. Crucially, LESS adapts existing influence formulations to work with the Adam optimizer and variable-length instruction data. LESS first constructs a highly reusable and transferable gradient datastore with low-dimensional gradient features and then selects examples based on their similarity to few-shot examples embodying a specific capability. Experiments show that training on a LESS-selected 5% of the data can often outperform training on the full dataset across diverse downstream tasks. Furthermore, the selected data is highly transferable: smaller models can be leveraged to select useful data for larger models and models from different families. Our qualitative analysis shows that our method goes beyond surface form cues to identify data that exemplifies the necessary reasoning skills for the intended downstream application.

http://arxiv.org/abs/2402.04333v2

cs.CL, cs.AI, cs.LG

Training Language Models to Generate Text with Citations via Fine-grained Rewards

Chengyu Huang, Zeqiu Wu, Yushi Hu, Wenya Wang

While recent Large Language Models (LLMs) have proven useful in answering user queries, they are prone to hallucination, and their responses often lack credibility due to missing references to reliable sources. An intuitive solution to these issues would be to include in-text citations referring to external documents as evidence. While previous works have directly prompted LLMs to generate in-text citations, their performances are far from satisfactory, especially when it comes to smaller LLMs. In this work, we propose an effective training framework using fine-grained rewards to teach LLMs to generate highly supportive and relevant citations, while ensuring the correctness of their responses. We also conduct a systematic analysis of applying these fine-grained rewards to common LLM training strategies, demonstrating its advantage over conventional practices. We conduct extensive experiments on Question Answering (QA) datasets taken from the ALCE benchmark and validate the model's generalizability using EXPERTQA. On LLaMA-2-7B, the incorporation of fine-grained rewards achieves the best performance among the baselines, even surpassing that of GPT-3.5-turbo.

http://arxiv.org/abs/2402.04315v1

cs.CL

Scaling Laws for Downstream Task Performance of Large Language Models

Berivan Isik, Natalia Ponomareva, Hussein Hazimeh, Dimitris Paparas, Sergei Vassilvitskii, Sanmi Koyejo

Scaling laws provide important insights that can guide the design of large language models (LLMs). Existing work has primarily focused on studying scaling laws for pretraining (upstream) loss. However, in transfer learning settings, in which LLMs are pretrained on an unsupervised dataset and then finetuned on a downstream task, we often also care about the downstream performance. In this work, we study the scaling behavior in a transfer learning setting, where LLMs are finetuned for machine translation tasks. Specifically, we investigate how the choice of the pretraining data and its size affect downstream performance (translation quality) as judged by two metrics: downstream cross-entropy and BLEU score. Our experiments indicate that the size of the finetuning dataset and the distribution alignment between the pretraining and downstream data significantly influence the scaling behavior. With sufficient alignment, both downstream cross-entropy and BLEU score improve monotonically with more pretraining data. In such cases, we show that it is possible to predict the downstream BLEU score with good accuracy using a log-law. However, there are also cases where moderate misalignment causes the BLEU score to fluctuate or get worse with more pretraining, whereas downstream cross-entropy monotonically improves. By analyzing these observations, we provide new practical insights for choosing appropriate pretraining data.

http://arxiv.org/abs/2402.04177v1

cs.CL, cs.LG, stat.ML

The Use of a Large Language Model for Cyberbullying Detection

Bayode Ogunleye, Babitha Dharmaraj

The dominance of social media has added to the channels of bullying for perpetrators. Unfortunately, cyberbullying (CB) is the most prevalent phenomenon in todays cyber world, and is a severe threat to the mental and physical health of citizens. This opens the need to develop a robust system to prevent bullying content from online forums, blogs, and social media platforms to manage the impact in our society. Several machine learning (ML) algorithms have been proposed for this purpose. However, their performances are not consistent due to high class imbalance and generalisation issues. In recent years, large language models (LLMs) like BERT and RoBERTa have achieved state-of-the-art (SOTA) results in several natural language processing (NLP) tasks. Unfortunately, the LLMs have not been applied extensively for CB detection. In our paper, we explored the use of these models for cyberbullying (CB) detection. We have prepared a new dataset (D2) from existing studies (Formspring and Twitter). Our experimental results for dataset D1 and D2 showed that RoBERTa outperformed other models.

http://arxiv.org/abs/2402.04088v1

cs.CL, cs.AI, cs.LG, stat.AP, H.3.3

Understanding the Effect of Noise in LLM Training Data with Algorithmic Chains of Thought

Alex Havrilla, Maia Iyer

During both pretraining and fine-tuning, Large Language Models (\textbf{LLMs}) are trained on trillions of tokens of text of widely varying quality. Both phases of training typically involve heuristically filtering out ``low-quality'' or \textit{noisy} training samples, yet little is known quantitatively about how the type or intensity of noise affects downstream performance. In this work, we study how noise in chain of thought (\textbf{CoT}) impacts task performance in the highly-controlled setting of algorithmically solvable tasks. First, we develop the Traced Integer (\textbf{TInt}) framework to generate highly customizable noised execution traces for any arithmetic function on lists of integers. We then define two types of noise: \textit{static} noise, a local form of noise which is applied after the CoT trace is computed, and \textit{dynamic} noise, a global form of noise which propagates errors in the trace as it is computed. We then evaluate the test performance of pretrained models both prompted and fine-tuned on noised datasets with varying levels of dataset contamination and intensity. We find fine-tuned models are extremely robust to high levels of static noise but struggle significantly more with lower levels of dynamic noise. In contrast, few-shot prompted models appear more sensitive to even static noise. We conclude with a discussion of how our findings impact noise filtering best-practices, in particular emphasizing the importance of removing samples containing destructive dynamic noise with global errors.

http://arxiv.org/abs/2402.04004v2

cs.LG

LV-Eval: A Balanced Long-Context Benchmark with 5 Length Levels Up to 256K

Tao Yuan, Xuefei Ning, Dong Zhou, Zhijie Yang, Shiyao Li, Minghui Zhuang, Zheyue Tan, Zhuyu Yao, Dahua Lin, Boxun Li, Guohao Dai, Shengen Yan, Yu Wang

State-of-the-art large language models (LLMs) are now claiming remarkable supported context lengths of 256k or even more. In contrast, the average context lengths of mainstream benchmarks are insufficient (5k-21k), and they suffer from potential knowledge leakage and inaccurate metrics, resulting in biased evaluation. This paper introduces LV-Eval, a challenging long-context benchmark with five length levels (16k, 32k, 64k, 128k, and 256k) reaching up to 256k words. LV-Eval features two main tasks, single-hop QA and multi-hop QA, comprising 11 bilingual datasets. The design of LV-Eval has incorporated three key techniques, namely confusing facts insertion, keyword and phrase replacement, and keyword-recall-based metric design. The advantages of LV-Eval include controllable evaluation across different context lengths, challenging test instances with confusing facts, mitigated knowledge leakage, and more objective evaluations. We evaluate 10 LLMs on LV-Eval and conduct ablation studies on the techniques used in LV-Eval construction. The results reveal that: (i) Commercial LLMs generally outperform open-source LLMs when evaluated within length levels shorter than their claimed context length. However, their overall performance is surpassed by open-source LLMs with longer context lengths. (ii) Extremely long-context LLMs, such as Yi-6B-200k, exhibit a relatively gentle degradation of performance, but their absolute performances may not necessarily be higher than those of LLMs with shorter context lengths. (iii) LLMs' performances can significantly degrade in the presence of confusing information, especially in the pressure test of "needle in a haystack". (iv) Issues related to knowledge leakage and inaccurate metrics introduce bias in evaluation, and these concerns are alleviated in LV-Eval. All datasets and evaluation codes are released at: https://github.com/infinigence/LVEval.

http://arxiv.org/abs/2402.05136v1

cs.CL

Can Large Language Models Detect Rumors on Social Media?

Qiang Liu, Xiang Tao, Junfei Wu, Shu Wu, Liang Wang

In this work, we investigate to use Large Language Models (LLMs) for rumor detection on social media. However, it is challenging for LLMs to reason over the entire propagation information on social media, which contains news contents and numerous comments, due to LLMs may not concentrate on key clues in the complex propagation information, and have trouble in reasoning when facing massive and redundant information. Accordingly, we propose an LLM-empowered Rumor Detection (LeRuD) approach, in which we design prompts to teach LLMs to reason over important clues in news and comments, and divide the entire propagation information into a Chain-of-Propagation for reducing LLMs' burden. We conduct extensive experiments on the Twitter and Weibo datasets, and LeRuD outperforms several state-of-the-art rumor detection models by 3.2% to 7.7%. Meanwhile, by applying LLMs, LeRuD requires no data for training, and thus shows more promising rumor detection ability in few-shot or zero-shot scenarios.

http://arxiv.org/abs/2402.03916v2

cs.IR, cs.CL

Rethinking Skill Extraction in the Job Market Domain using Large Language Models

Khanh Cao Nguyen, Mike Zhang, Syrielle Montariol, Antoine Bosselut

Skill Extraction involves identifying skills and qualifications mentioned in documents such as job postings and resumes. The task is commonly tackled by training supervised models using a sequence labeling approach with BIO tags. However, the reliance on manually annotated data limits the generalizability of such approaches. Moreover, the common BIO setting limits the ability of the models to capture complex skill patterns and handle ambiguous mentions. In this paper, we explore the use of in-context learning to overcome these challenges, on a benchmark of 6 uniformized skill extraction datasets. Our approach leverages the few-shot learning capabilities of large language models (LLMs) to identify and extract skills from sentences. We show that LLMs, despite not being on par with traditional supervised models in terms of performance, can better handle syntactically complex skill mentions in skill extraction tasks.

http://arxiv.org/abs/2402.03832v1

cs.CL

MolTC: Towards Molecular Relational Modeling In Language Models

Junfeng Fang, Shuai Zhang, Chang Wu, Zhengyi Yang, Zhiyuan Liu, Sihang Li, Kun Wang, Wenjie Du, Xiang Wang

Molecular Relational Learning (MRL), aiming to understand interactions between molecular pairs, plays a pivotal role in advancing biochemical research. Recently, the adoption of large language models (LLMs), known for their vast knowledge repositories and advanced logical inference capabilities, has emerged as a promising way for efficient and effective MRL. Despite their potential, these methods predominantly rely on the textual data, thus not fully harnessing the wealth of structural information inherent in molecular graphs. Moreover, the absence of a unified framework exacerbates the issue of information underutilization, as it hinders the sharing of interaction mechanism learned across diverse datasets. To address these challenges, this work proposes a novel LLM-based multi-modal framework for Molecular inTeraction prediction following Chain-of-Thought (CoT) theory, termed MolTC, which effectively integrate graphical information of two molecules in pair. For achieving a unified MRL, MolTC innovatively develops a dynamic parameter-sharing strategy for cross-dataset information sharing. Moreover, to train MolTC efficiently, we introduce a Multi-hierarchical CoT concept to refine its training paradigm, and conduct a comprehensive Molecular Interactive Instructions dataset for the development of biochemical LLMs involving MRL. Our experiments, conducted across various datasets involving over 4,000,000 molecular pairs, exhibit the superiority of our method over current GNN and LLM-based baselines. Code is available at https://github.com/MangoKiller/MolTC.

http://arxiv.org/abs/2402.03781v5

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

Tuning Large Multimodal Models for Videos using Reinforcement Learning from AI Feedback

Daechul Ahn, Yura Choi, Youngjae Yu, Dongyeop Kang, Jonghyun Choi

Recent advancements in large language models have influenced the development of video large multimodal models (VLMMs). The previous approaches for VLMMs involved Supervised Fine-Tuning (SFT) with instruction-tuned datasets, integrating LLM with visual encoders, and adding additional learnable modules. Video and text multimodal alignment remains challenging, primarily due to the deficient volume and quality of multimodal instruction-tune data compared to text-only data. We present a novel alignment strategy that employs multimodal AI system to oversee itself called Reinforcement Learning from AI Feedback (RLAIF), providing self-preference feedback to refine itself and facilitating the alignment of video and text modalities. In specific, we propose context-aware reward modeling by providing detailed video descriptions as context during the generation of preference feedback in order to enrich the understanding of video content. Demonstrating enhanced performance across diverse video benchmarks, our multimodal RLAIF approach, VLM-RLAIF, outperforms existing approaches, including the SFT model. We commit to open-sourcing our code, models, and datasets to foster further research in this area.

http://arxiv.org/abs/2402.03746v2

cs.CV

Similarity-based Neighbor Selection for Graph LLMs

Rui Li, Jiwei Li, Jiawei Han, Guoyin Wang

Text-attributed graphs (TAGs) present unique challenges for direct processing by Language Learning Models (LLMs), yet their extensive commonsense knowledge and robust reasoning capabilities offer great promise for node classification in TAGs. Prior research in this field has grappled with issues such as over-squashing, heterophily, and ineffective graph information integration, further compounded by inconsistencies in dataset partitioning and underutilization of advanced LLMs. To address these challenges, we introduce Similarity-based Neighbor Selection (SNS). Using SimCSE and advanced neighbor selection techniques, SNS effectively improves the quality of selected neighbors, thereby improving graph representation and alleviating issues like over-squashing and heterophily. Besides, as an inductive and training-free approach, SNS demonstrates superior generalization and scalability over traditional GNN methods. Our comprehensive experiments, adhering to standard dataset partitioning practices, demonstrate that SNS, through simple prompt interactions with LLMs, consistently outperforms vanilla GNNs and achieves state-of-the-art results on datasets like PubMed in node classification, showcasing LLMs' potential in graph structure understanding. Our research further underscores the significance of graph structure integration in LLM applications and identifies key factors for their success in node classification. Code is available at https://github.com/ruili33/SNS.

http://arxiv.org/abs/2402.03720v1

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

Empowering Language Models with Active Inquiry for Deeper Understanding

Jing-Cheng Pang, Heng-Bo Fan, Pengyuan Wang, Jia-Hao Xiao, Nan Tang, Si-Hang Yang, Chengxing Jia, Sheng-Jun Huang, Yang Yu

The rise of large language models (LLMs) has revolutionized the way that we interact with artificial intelligence systems through natural language. However, LLMs often misinterpret user queries because of their uncertain intention, leading to less helpful responses. In natural human interactions, clarification is sought through targeted questioning to uncover obscure information. Thus, in this paper, we introduce LaMAI (Language Model with Active Inquiry), designed to endow LLMs with this same level of interactive engagement. LaMAI leverages active learning techniques to raise the most informative questions, fostering a dynamic bidirectional dialogue. This approach not only narrows the contextual gap but also refines the output of the LLMs, aligning it more closely with user expectations. Our empirical studies, across a variety of complex datasets where LLMs have limited conversational context, demonstrate the effectiveness of LaMAI. The method improves answer accuracy from 31.9% to 50.9%, outperforming other leading question-answering frameworks. Moreover, in scenarios involving human participants, LaMAI consistently generates responses that are superior or comparable to baseline methods in more than 82% of the cases. The applicability of LaMAI is further evidenced by its successful integration with various LLMs, highlighting its potential for the future of interactive language models.

http://arxiv.org/abs/2402.03719v1

cs.CL, cs.AI

Are Machines Better at Complex Reasoning? Unveiling Human-Machine Inference Gaps in Entailment Verification

Soumya Sanyal, Tianyi Xiao, Jiacheng Liu, Wenya Wang, Xiang Ren

Making inferences in text comprehension to understand the meaning is essential in language processing. This work studies the entailment verification (EV) problem of multi-sentence premises that requires a system to make multiple inferences implicitly. Studying EV for such complex premises is important because modern NLP problems, such as detecting inconsistent model-generated rationales, require complex multi-hop reasoning. However, current textual inference datasets mostly contain short premises that only partially focus on these challenges. To address this, we compile an EV benchmark that includes datasets from three NLP domains (NLI, contextual QA, and rationales) containing multi-sentence premises. On benchmarking humans and LLMs, we find that LLMs are better than humans in multi-hop reasoning across extended contexts, while humans perform better in simple deductive reasoning tasks. We also finetune a Flan-T5 model for EV using two training objectives to obtain a strong open-source model that outperforms GPT-3.5 and rivals GPT-4. Finally, we use this model to filter out inconsistent model-generated rationales in self-consistency decoding, resulting in a 6% accuracy improvement on average across three MCQ datasets.

http://arxiv.org/abs/2402.03686v2

cs.CL, cs.AI

Resolving Transcription Ambiguity in Spanish: A Hybrid Acoustic-Lexical System for Punctuation Restoration

Xiliang Zhu, Chia-Tien Chang, Shayna Gardiner, David Rossouw, Jonas Robertson

Punctuation restoration is a crucial step after Automatic Speech Recognition (ASR) systems to enhance transcript readability and facilitate subsequent NLP tasks. Nevertheless, conventional lexical-based approaches are inadequate for solving the punctuation restoration task in Spanish, where ambiguity can be often found between unpunctuated declaratives and questions. In this study, we propose a novel hybrid acoustic-lexical punctuation restoration system for Spanish transcription, which consolidates acoustic and lexical signals through a modular process. Our experiment results show that the proposed system can effectively improve F1 score of question marks and overall punctuation restoration on both public and internal Spanish conversational datasets. Additionally, benchmark comparison against LLMs (Large Language Model) indicates the superiority of our approach in accuracy, reliability and latency. Furthermore, we demonstrate that the Word Error Rate (WER) of the ASR module also benefits from our proposed system.

http://arxiv.org/abs/2402.03519v1

cs.CL, cs.AI