title
stringlengths 26
159
| content
stringlengths 365
1.97k
| classification
stringclasses 2
values |
|---|---|---|
Quantifying In-Context Reasoning Effects and Memorization Effects in LLMs | arXiv:2405.11880v1 Announce Type: cross
Abstract: In this study, we propose an axiomatic system to define and quantify the precise memorization and in-context reasoning effects used by the large language model (LLM) for language generation. These effects are formulated as non-linear interactions between tokens/words encoded by the LLM. Specifically, the axiomatic system enables us to categorize the memorization effects into foundational memorization effects and chaotic memorization effects, and further classify in-context reasoning effects into enhanced inference patterns, eliminated inference patterns, and reversed inference patterns. Besides, the decomposed effects satisfy the sparsity property and the universal matching property, which mathematically guarantee that the LLM's confidence score can be faithfully decomposed into the memorization effects and in-context reasoning effects. Experiments show that the clear disentanglement of memorization effects and in-context reasoning effects enables a straightforward examination of detailed inference patterns encoded by LLMs. | natural language processing |
On Efficient and Statistical Quality Estimation for Data Annotation | arXiv:2405.11919v1 Announce Type: cross
Abstract: Annotated datasets are an essential ingredient to train, evaluate, compare and productionalize supervised machine learning models. It is therefore imperative that annotations are of high quality. For their creation, good quality management and thereby reliable quality estimates are needed. Then, if quality is insufficient during the annotation process, rectifying measures can be taken to improve it. Quality estimation is often performed by having experts manually label instances as correct or incorrect. But checking all annotated instances tends to be expensive. Therefore, in practice, usually only subsets are inspected; sizes are chosen mostly without justification or regard to statistical power and more often than not, are relatively small. Basing estimates on small sample sizes, however, can lead to imprecise values for the error rate. Using unnecessarily large sample sizes costs money that could be better spent, for instance on more annotations. Therefore, we first describe in detail how to use confidence intervals for finding the minimal sample size needed to estimate the annotation error rate. Then, we propose applying acceptance sampling as an alternative to error rate estimation We show that acceptance sampling can reduce the required sample sizes up to 50% while providing the same statistical guarantees. | natural language processing |
KG-RAG: Bridging the Gap Between Knowledge and Creativity | arXiv:2405.12035v1 Announce Type: cross
Abstract: Ensuring factual accuracy while maintaining the creative capabilities of Large Language Model Agents (LMAs) poses significant challenges in the development of intelligent agent systems. LMAs face prevalent issues such as information hallucinations, catastrophic forgetting, and limitations in processing long contexts when dealing with knowledge-intensive tasks. This paper introduces a KG-RAG (Knowledge Graph-Retrieval Augmented Generation) pipeline, a novel framework designed to enhance the knowledge capabilities of LMAs by integrating structured Knowledge Graphs (KGs) with the functionalities of LLMs, thereby significantly reducing the reliance on the latent knowledge of LLMs. The KG-RAG pipeline constructs a KG from unstructured text and then performs information retrieval over the newly created graph to perform KGQA (Knowledge Graph Question Answering). The retrieval methodology leverages a novel algorithm called Chain of Explorations (CoE) which benefits from LLMs reasoning to explore nodes and relationships within the KG sequentially. Preliminary experiments on the ComplexWebQuestions dataset demonstrate notable improvements in the reduction of hallucinated content and suggest a promising path toward developing intelligent systems adept at handling knowledge-intensive tasks. | natural language processing |
Imp: Highly Capable Large Multimodal Models for Mobile Devices | arXiv:2405.12107v1 Announce Type: cross
Abstract: By harnessing the capabilities of large language models (LLMs), recent large multimodal models (LMMs) have shown remarkable versatility in open-world multimodal understanding. Nevertheless, they are usually parameter-heavy and computation-intensive, thus hindering their applicability in resource-constrained scenarios. To this end, several lightweight LMMs have been proposed successively to maximize the capabilities under constrained scale (e.g., 3B). Despite the encouraging results achieved by these methods, most of them only focus on one or two aspects of the design space, and the key design choices that influence model capability have not yet been thoroughly investigated. In this paper, we conduct a systematic study for lightweight LMMs from the aspects of model architecture, training strategy, and training data. Based on our findings, we obtain Imp -- a family of highly capable LMMs at the 2B-4B scales. Notably, our Imp-3B model steadily outperforms all the existing lightweight LMMs of similar size, and even surpasses the state-of-the-art LMMs at the 13B scale. With low-bit quantization and resolution reduction techniques, our Imp model can be deployed on a Qualcomm Snapdragon 8Gen3 mobile chip with a high inference speed of about 13 tokens/s. | natural language processing |
Reindex-Then-Adapt: Improving Large Language Models for Conversational Recommendation | arXiv:2405.12119v1 Announce Type: cross
Abstract: Large language models (LLMs) are revolutionizing conversational recommender systems by adeptly indexing item content, understanding complex conversational contexts, and generating relevant item titles. However, controlling the distribution of recommended items remains a challenge. This leads to suboptimal performance due to the failure to capture rapidly changing data distributions, such as item popularity, on targeted conversational recommendation platforms. In conversational recommendation, LLMs recommend items by generating the titles (as multiple tokens) autoregressively, making it difficult to obtain and control the recommendations over all items. Thus, we propose a Reindex-Then-Adapt (RTA) framework, which converts multi-token item titles into single tokens within LLMs, and then adjusts the probability distributions over these single-token item titles accordingly. The RTA framework marries the benefits of both LLMs and traditional recommender systems (RecSys): understanding complex queries as LLMs do; while efficiently controlling the recommended item distributions in conversational recommendations as traditional RecSys do. Our framework demonstrates improved accuracy metrics across three different conversational recommendation datasets and two adaptation settings | natural language processing |
Eliciting Problem Specifications via Large Language Models | arXiv:2405.12147v1 Announce Type: cross
Abstract: Cognitive systems generally require a human to translate a problem definition into some specification that the cognitive system can use to attempt to solve the problem or perform the task. In this paper, we illustrate that large language models (LLMs) can be utilized to map a problem class, defined in natural language, into a semi-formal specification that can then be utilized by an existing reasoning and learning system to solve instances from the problem class. We present the design of LLM-enabled cognitive task analyst agent(s). Implemented with LLM agents, this system produces a definition of problem spaces for tasks specified in natural language. LLM prompts are derived from the definition of problem spaces in the AI literature and general problem-solving strategies (Polya's How to Solve It). A cognitive system can then use the problem-space specification, applying domain-general problem solving strategies ("weak methods" such as search), to solve multiple instances of problems from the problem class. This result, while preliminary, suggests the potential for speeding cognitive systems research via disintermediation of problem formulation while also retaining core capabilities of cognitive systems, such as robust inference and online learning. | natural language processing |
IT5: Text-to-text Pretraining for Italian Language Understanding and Generation | arXiv:2203.03759v2 Announce Type: replace
Abstract: We introduce IT5, the first family of encoder-decoder transformer models pretrained specifically on Italian. We document and perform a thorough cleaning procedure for a large Italian corpus and use it to pretrain four IT5 model sizes. We then introduce the ItaGen benchmark, which includes a broad range of natural language understanding and generation tasks for Italian, and use it to evaluate the performance of IT5 models and multilingual baselines. We find monolingual IT5 models to provide the best scale-to-performance ratio across tested models, consistently outperforming their multilingual counterparts and setting a new state-of-the-art for Italian language generation. | natural language processing |
Acquiring and Modelling Abstract Commonsense Knowledge via Conceptualization | arXiv:2206.01532v2 Announce Type: replace
Abstract: Conceptualization, or viewing entities and situations as instances of abstract concepts in mind and making inferences based on that, is a vital component in human intelligence for commonsense reasoning. Despite recent progress in artificial intelligence to acquire and model commonsense attributed to neural language models and commonsense knowledge graphs (CKGs), conceptualization is yet to be introduced thoroughly, making current approaches ineffective to cover knowledge about countless diverse entities and situations in the real world.
To address the problem, we thoroughly study the role of conceptualization in commonsense reasoning, and formulate a framework to replicate human conceptual induction by acquiring abstract knowledge about events regarding abstract concepts, as well as higher-level triples or inferences upon them. We then apply the framework to ATOMIC, a large-scale human-annotated CKG, aided by the taxonomy Probase. We annotate a dataset on the validity of contextualized conceptualizations from ATOMIC on both event and triple levels, develop a series of heuristic rules based on linguistic features, and train a set of neural models to generate and verify abstract knowledge. Based on these components, a pipeline to acquire abstract knowledge is built. A large abstract CKG upon ATOMIC is then induced, ready to be instantiated to infer about unseen entities or situations. Finally, we empirically show the benefits of augmenting CKGs with abstract knowledge in downstream tasks like commonsense inference and zero-shot commonsense QA. | natural language processing |
Knowledge-augmented Graph Neural Networks with Concept-aware Attention for Adverse Drug Event Detection | arXiv:2301.10451v3 Announce Type: replace
Abstract: Adverse drug events (ADEs) are an important aspect of drug safety. Various texts such as biomedical literature, drug reviews, and user posts on social media and medical forums contain a wealth of information about ADEs. Recent studies have applied word embedding and deep learning -based natural language processing to automate ADE detection from text. However, they did not explore incorporating explicit medical knowledge about drugs and adverse reactions or the corresponding feature learning. This paper adopts the heterogenous text graph which describes relationships between documents, words and concepts, augments it with medical knowledge from the Unified Medical Language System, and proposes a concept-aware attention mechanism which learns features differently for the different types of nodes in the graph. We further utilize contextualized embeddings from pretrained language models and convolutional graph neural networks for effective feature representation and relational learning. Experiments on four public datasets show that our model achieves performance competitive to the recent advances and the concept-aware attention consistently outperforms other attention mechanisms. | natural language processing |
Multimodal Chain-of-Thought Reasoning in Language Models | arXiv:2302.00923v5 Announce Type: replace
Abstract: Large language models (LLMs) have shown impressive performance on complex reasoning by leveraging chain-of-thought (CoT) prompting to generate intermediate reasoning chains as the rationale to infer the answer. However, existing CoT studies have primarily focused on the language modality. We propose Multimodal-CoT that incorporates language (text) and vision (images) modalities into a two-stage framework that separates rationale generation and answer inference. In this way, answer inference can leverage better generated rationales that are based on multimodal information. Experimental results on ScienceQA and A-OKVQA benchmark datasets show the effectiveness of our proposed approach. With Multimodal-CoT, our model under 1 billion parameters achieves state-of-the-art performance on the ScienceQA benchmark. Our analysis indicates that Multimodal-CoT offers the advantages of mitigating hallucination and enhancing convergence speed. Code is publicly available at https://github.com/amazon-science/mm-cot. | computer vision |
COKE: A Cognitive Knowledge Graph for Machine Theory of Mind | arXiv:2305.05390v2 Announce Type: replace
Abstract: Theory of mind (ToM) refers to humans' ability to understand and infer the desires, beliefs, and intentions of others. The acquisition of ToM plays a key role in humans' social cognition and interpersonal relations. Though indispensable for social intelligence, ToM is still lacking for modern AI and NLP systems since they cannot access the human mental state and cognitive process beneath the training corpus. To empower AI systems with the ToM ability and narrow the gap between them and humans, in this paper, we propose COKE: the first cognitive knowledge graph for machine theory of mind. Specifically, COKE formalizes ToM as a collection of 45k+ manually verified cognitive chains that characterize human mental activities and subsequent behavioral/affective responses when facing specific social circumstances. In addition, we further generalize COKE using LLMs and build a powerful generation model COLM tailored for cognitive reasoning. Experimental results in both automatic and human evaluation demonstrate the high quality of COKE, the superior ToM ability of COLM, and its potential to significantly enhance social applications. | natural language processing |
Deepfake Text Detection in the Wild | arXiv:2305.13242v2 Announce Type: replace
Abstract: Large language models (LLMs) have achieved human-level text generation, emphasizing the need for effective AI-generated text detection to mitigate risks like the spread of fake news and plagiarism. Existing research has been constrained by evaluating detection methods on specific domains or particular language models. In practical scenarios, however, the detector faces texts from various domains or LLMs without knowing their sources. To this end, we build a comprehensive testbed by gathering texts from diverse human writings and texts generated by different LLMs. Empirical results show challenges in distinguishing machine-generated texts from human-authored ones across various scenarios, especially out-of-distribution. These challenges are due to the decreasing linguistic distinctions between the two sources. Despite challenges, the top-performing detector can identify 86.54% out-of-domain texts generated by a new LLM, indicating the feasibility for application scenarios. We release our resources at https://github.com/yafuly/MAGE. | natural language processing |
Generating with Confidence: Uncertainty Quantification for Black-box Large Language Models | arXiv:2305.19187v3 Announce Type: replace
Abstract: Large language models (LLMs) specializing in natural language generation (NLG) have recently started exhibiting promising capabilities across a variety of domains. However, gauging the trustworthiness of responses generated by LLMs remains an open challenge, with limited research on uncertainty quantification (UQ) for NLG. Furthermore, existing literature typically assumes white-box access to language models, which is becoming unrealistic either due to the closed-source nature of the latest LLMs or computational constraints. In this work, we investigate UQ in NLG for *black-box* LLMs. We first differentiate *uncertainty* vs *confidence*: the former refers to the ``dispersion'' of the potential predictions for a fixed input, and the latter refers to the confidence on a particular prediction/generation. We then propose and compare several confidence/uncertainty measures, applying them to *selective NLG* where unreliable results could either be ignored or yielded for further assessment. Experiments were carried out with several popular LLMs on question-answering datasets (for evaluation purposes). Results reveal that a simple measure for the semantic dispersion can be a reliable predictor of the quality of LLM responses, providing valuable insights for practitioners on uncertainty management when adopting LLMs. The code to replicate our experiments is available at https://github.com/zlin7/UQ-NLG. | natural language processing |
MultiLegalPile: A 689GB Multilingual Legal Corpus | arXiv:2306.02069v3 Announce Type: replace
Abstract: Large, high-quality datasets are crucial for training Large Language Models (LLMs). However, so far, there are few datasets available for specialized critical domains such as law and the available ones are often only for the English language. We curate and release MultiLegalPile, a 689GB corpus in 24 languages from 17 jurisdictions. The MultiLegalPile corpus, which includes diverse legal data sources with varying licenses, allows for pretraining NLP models under fair use, with more permissive licenses for the Eurlex Resources and Legal mC4 subsets. We pretrain two RoBERTa models and one Longformer multilingually, and 24 monolingual models on each of the language-specific subsets and evaluate them on LEXTREME. Additionally, we evaluate the English and multilingual models on LexGLUE. Our multilingual models set a new SotA on LEXTREME and our English models on LexGLUE. We release the dataset, the trained models, and all of the code under the most open possible licenses. | natural language processing |
Controllable Data Augmentation for Few-Shot Text Mining with Chain-of-Thought Attribute Manipulation | arXiv:2307.07099v2 Announce Type: replace
Abstract: Prompting large language models (LLMs) for data augmentation has recently become a common practice in few-shot NLP tasks. In this paper, we propose Chain-of-Thought Attribute Manipulation (CoTAM), a novel approach that generates new data from existing examples by only tweaking in the user-provided, task-specific attribute, e.g., sentiment polarity or topic in movie reviews. Instead of conventional latent representation controlling, we leverage the chain-of-thought prompting to directly edit the text in three steps, (1) attribute decomposition, (2) manipulation proposal, and (3) sentence reconstruction. Extensive results on various tasks, such as text (pair) classification, aspect-based sentiment analysis, and conditional text generation, verify the superiority of CoTAM over other LLM-based augmentation methods with the same number of training examples for both fine-tuning and in-context learning. Remarkably, the 2D visualization of the augmented dataset using principal component analysis revealed a human-recognizable decision boundary that is likely hinted by the attribute manipulation, demonstrating the potential of our proposed approach. | natural language processing |
NaijaRC: A Multi-choice Reading Comprehension Dataset for Nigerian Languages | arXiv:2308.09768v3 Announce Type: replace
Abstract: In this paper, we create NaijaRC: a new multi-choice Reading Comprehension dataset for three native Nigeria languages that is based on high-school reading comprehension examination. We provide baseline results by performing cross-lingual transfer using existing English RACE and Belebele training dataset based on a pre-trained encoder-only model. Additionally, we provide results by prompting large language models (LLMs) like GPT-4. | natural language processing |
Anonymity at Risk? Assessing Re-Identification Capabilities of Large Language Models | arXiv:2308.11103v2 Announce Type: replace
Abstract: Anonymity of both natural and legal persons in court rulings is a critical aspect of privacy protection in the European Union and Switzerland. With the advent of LLMs, concerns about large-scale re-identification of anonymized persons are growing. In accordance with the Federal Supreme Court of Switzerland, we explore the potential of LLMs to re-identify individuals in court rulings by constructing a proof-of-concept using actual legal data from the Swiss federal supreme court. Following the initial experiment, we constructed an anonymized Wikipedia dataset as a more rigorous testing ground to further investigate the findings. With the introduction and application of the new task of re-identifying people in texts, we also introduce new metrics to measure performance. We systematically analyze the factors that influence successful re-identifications, identifying model size, input length, and instruction tuning among the most critical determinants. Despite high re-identification rates on Wikipedia, even the best LLMs struggled with court decisions. The complexity is attributed to the lack of test datasets, the necessity for substantial training resources, and data sparsity in the information used for re-identification. In conclusion, this study demonstrates that re-identification using LLMs may not be feasible for now, but as the proof-of-concept on Wikipedia showed, it might become possible in the future. We hope that our system can help enhance the confidence in the security of anonymized decisions, thus leading to the courts being more confident to publish decisions. | natural language processing |
Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding | arXiv:2309.08168v2 Announce Type: replace
Abstract: We present a novel inference scheme, self-speculative decoding, for accelerating Large Language Models (LLMs) without the need for an auxiliary model. This approach is characterized by a two-stage process: drafting and verification. The drafting stage generates draft tokens at a slightly lower quality but more quickly, which is achieved by selectively skipping certain intermediate layers during drafting. Subsequently, the verification stage employs the original LLM to validate those draft output tokens in one forward pass. This process ensures the final output remains identical to that produced by the unaltered LLM. Moreover, the proposed method requires no additional neural network training and no extra memory footprint, making it a plug-and-play and cost-effective solution for inference acceleration. Benchmarks with LLaMA-2 and its variants demonstrated a speedup up to 1.99$\times$. | natural language processing |
Evaluating Gender Bias of Pre-trained Language Models in Natural Language Inference by Considering All Labels | arXiv:2309.09697v3 Announce Type: replace
Abstract: Discriminatory gender biases have been found in Pre-trained Language Models (PLMs) for multiple languages. In Natural Language Inference (NLI), existing bias evaluation methods have focused on the prediction results of one specific label out of three labels, such as neutral. However, such evaluation methods can be inaccurate since unique biased inferences are associated with unique prediction labels. Addressing this limitation, we propose a bias evaluation method for PLMs, called NLI-CoAL, which considers all the three labels of NLI task. First, we create three evaluation data groups that represent different types of biases. Then, we define a bias measure based on the corresponding label output of each data group. In the experiments, we introduce a meta-evaluation technique for NLI bias measures and use it to confirm that our bias measure can distinguish biased, incorrect inferences from non-biased incorrect inferences better than the baseline, resulting in a more accurate bias evaluation. We create the datasets in English, Japanese, and Chinese, and successfully validate the compatibility of our bias measure across multiple languages. Lastly, we observe the bias tendencies in PLMs of different languages. To our knowledge, we are the first to construct evaluation datasets and measure PLMs' bias from NLI in Japanese and Chinese. | natural language processing |
K-pop Lyric Translation: Dataset, Analysis, and Neural-Modelling | arXiv:2309.11093v4 Announce Type: replace
Abstract: Lyric translation, a field studied for over a century, is now attracting computational linguistics researchers. We identified two limitations in previous studies. Firstly, lyric translation studies have predominantly focused on Western genres and languages, with no previous study centering on K-pop despite its popularity. Second, the field of lyric translation suffers from a lack of publicly available datasets; to the best of our knowledge, no such dataset exists. To broaden the scope of genres and languages in lyric translation studies, we introduce a novel singable lyric translation dataset, approximately 89\% of which consists of K-pop song lyrics. This dataset aligns Korean and English lyrics line-by-line and section-by-section. We leveraged this dataset to unveil unique characteristics of K-pop lyric translation, distinguishing it from other extensively studied genres, and to construct a neural lyric translation model, thereby underscoring the importance of a dedicated dataset for singable lyric translations. | natural language processing |
You Only Look at Screens: Multimodal Chain-of-Action Agents | arXiv:2309.11436v3 Announce Type: replace
Abstract: Autonomous graphical user interface (GUI) agents aim to facilitate task automation by interacting with the user interface without manual intervention. Recent studies have investigated eliciting the capabilities of large language models (LLMs) for effective engagement in diverse environments. To align with the input-output requirement of LLMs, most existing approaches are developed under a sandbox setting where they rely on external tools and application-specific APIs to parse the environment into textual elements and interpret the predicted actions. Consequently, those approaches often grapple with inference inefficiency and error propagation risks. To mitigate the challenges, we introduce Auto-GUI, a multimodal solution that directly interacts with the interface, bypassing the need for environment parsing or reliance on application-dependent APIs. Moreover, we propose a chain-of-action technique -- leveraging a series of intermediate previous action histories and future action plans -- to help the agent decide what action to execute. We evaluate our approach on a new device-control benchmark AITW with 30$K$ unique instructions, spanning multi-step tasks such as application operation, web searching, and web shopping. Experimental results show that Auto-GUI achieves state-of-the-art performance with an action type prediction accuracy of 90\% and an overall action success rate of 74\%. Code is publicly available at https://github.com/cooelf/Auto-GUI. | computer vision |
Jury: A Comprehensive Evaluation Toolkit | arXiv:2310.02040v2 Announce Type: replace
Abstract: Evaluation plays a critical role in deep learning as a fundamental block of any prediction-based system. However, the vast number of Natural Language Processing (NLP) tasks and the development of various metrics have led to challenges in evaluating different systems with different metrics. To address these challenges, we introduce jury, a toolkit that provides a unified evaluation framework with standardized structures for performing evaluation across different tasks and metrics. The objective of jury is to standardize and improve metric evaluation for all systems and aid the community in overcoming the challenges in evaluation. Since its open-source release, jury has reached a wide audience and is available at https://github.com/obss/jury. | natural language processing |
Direct Neural Machine Translation with Task-level Mixture of Experts models | arXiv:2310.12236v2 Announce Type: replace
Abstract: Direct neural machine translation (direct NMT) is a type of NMT system that translates text between two non-English languages. Direct NMT systems often face limitations due to the scarcity of parallel data between non-English language pairs. Several approaches have been proposed to address this limitation, such as multilingual NMT and pivot NMT (translation between two languages via English). Task-level Mixture of expert models (Task-level MoE), an inference-efficient variation of Transformer-based models, has shown promising NMT performance for a large number of language pairs. In Task-level MoE, different language groups can use different routing strategies to optimize cross-lingual learning and inference speed. In this work, we examine Task-level MoE's applicability in direct NMT and propose a series of high-performing training and evaluation configurations, through which Task-level MoE-based direct NMT systems outperform bilingual and pivot-based models for a large number of low and high-resource direct pairs, and translation directions. Our Task-level MoE with 16 experts outperforms bilingual NMT, Pivot NMT models for 7 language pairs, while pivot-based models still performed better in 9 pairs and directions. | natural language processing |
Natural Language Interfaces for Tabular Data Querying and Visualization: A Survey | arXiv:2310.17894v3 Announce Type: replace
Abstract: The emergence of natural language processing has revolutionized the way users interact with tabular data, enabling a shift from traditional query languages and manual plotting to more intuitive, language-based interfaces. The rise of large language models (LLMs) such as ChatGPT and its successors has further advanced this field, opening new avenues for natural language processing techniques. This survey presents a comprehensive overview of natural language interfaces for tabular data querying and visualization, which allow users to interact with data using natural language queries. We introduce the fundamental concepts and techniques underlying these interfaces with a particular emphasis on semantic parsing, the key technology facilitating the translation from natural language to SQL queries or data visualization commands. We then delve into the recent advancements in Text-to-SQL and Text-to-Vis problems from the perspectives of datasets, methodologies, metrics, and system designs. This includes a deep dive into the influence of LLMs, highlighting their strengths, limitations, and potential for future improvements. Through this survey, we aim to provide a roadmap for researchers and practitioners interested in developing and applying natural language interfaces for data interaction in the era of large language models. | natural language processing |
Using Natural Language Explanations to Improve Robustness of In-context Learning | arXiv:2311.07556v2 Announce Type: replace
Abstract: Recent studies demonstrated that large language models (LLMs) can excel in many tasks via in-context learning (ICL). However, recent works show that ICL-prompted models tend to produce inaccurate results when presented with adversarial inputs. In this work, we investigate whether augmenting ICL with natural language explanations (NLEs) improves the robustness of LLMs on adversarial datasets covering natural language inference and paraphrasing identification. We prompt LLMs with a small set of human-generated NLEs to produce further NLEs, yielding more accurate results than both a zero-shot-ICL setting and using only human-generated NLEs. Our results on five popular LLMs (GPT3.5-turbo, Llama2, Vicuna, Zephyr, and Mistral) show that our approach yields over 6% improvement over baseline approaches for eight adversarial datasets: HANS, ISCS, NaN, ST, PICD, PISP, ANLI, and PAWS. Furthermore, previous studies have demonstrated that prompt selection strategies significantly enhance ICL on in-distribution test sets. However, our findings reveal that these strategies do not match the efficacy of our approach for robustness evaluations, resulting in an accuracy drop of 8% compared to the proposed approach. | natural language processing |
Deception Detection from Linguistic and Physiological Data Streams Using Bimodal Convolutional Neural Networks | arXiv:2311.10944v3 Announce Type: replace
Abstract: Deception detection is gaining increasing interest due to ethical and security concerns. This paper explores the application of convolutional neural networks for the purpose of multimodal deception detection. We use a dataset built by interviewing 104 subjects about two topics, with one truthful and one falsified response from each subject about each topic. In particular, we make three main contributions. First, we extract linguistic and physiological features from this data to train and construct the neural network models. Second, we propose a fused convolutional neural network model using both modalities in order to achieve an improved overall performance. Third, we compare our new approach with earlier methods designed for multimodal deception detection. We find that our system outperforms regular classification methods; our results indicate the feasibility of using neural networks for deception detection even in the presence of limited amounts of data. | natural language processing |
Learn or Recall? Revisiting Incremental Learning with Pre-trained Language Models | arXiv:2312.07887v2 Announce Type: replace
Abstract: Incremental Learning (IL) has been a long-standing problem in both vision and Natural Language Processing (NLP) communities. In recent years, as Pre-trained Language Models (PLMs) have achieved remarkable progress in various NLP downstream tasks, utilizing PLMs as backbones has become a common practice in recent research of IL in NLP. Most assume that catastrophic forgetting is the biggest obstacle to achieving superior IL performance and propose various techniques to overcome this issue. However, we find that this assumption is problematic. Specifically, we revisit more than 20 methods on four classification tasks (Text Classification, Intent Classification, Relation Extraction, and Named Entity Recognition) under the two most popular IL settings (Class-Incremental and Task-Incremental) and reveal that most of them severely underestimate the inherent anti-forgetting ability of PLMs. Based on the observation, we propose a frustratingly easy method called SEQ* for IL with PLMs. The results show that SEQ* has competitive or superior performance compared to state-of-the-art (SOTA) IL methods and requires considerably less trainable parameters and training time. These findings urge us to revisit the IL with PLMs and encourage future studies to have a fundamental understanding of the catastrophic forgetting in PLMs. The data, code and scripts are publicly available at https://github.com/zzz47zzz/pretrained-lm-for-incremental-learning. | natural language processing |
MoSECroT: Model Stitching with Static Word Embeddings for Crosslingual Zero-shot Transfer | arXiv:2401.04821v2 Announce Type: replace
Abstract: Transformer-based pre-trained language models (PLMs) have achieved remarkable performance in various natural language processing (NLP) tasks. However, pre-training such models can take considerable resources that are almost only available to high-resource languages. On the contrary, static word embeddings are easier to train in terms of computing resources and the amount of data required. In this paper, we introduce MoSECroT Model Stitching with Static Word Embeddings for Crosslingual Zero-shot Transfer), a novel and challenging task that is especially relevant to low-resource languages for which static word embeddings are available. To tackle the task, we present the first framework that leverages relative representations to construct a common space for the embeddings of a source language PLM and the static word embeddings of a target language. In this way, we can train the PLM on source-language training data and perform zero-shot transfer to the target language by simply swapping the embedding layer. However, through extensive experiments on two classification datasets, we show that although our proposed framework is competitive with weak baselines when addressing MoSECroT, it fails to achieve competitive results compared with some strong baselines. In this paper, we attempt to explain this negative result and provide several thoughts on possible improvement. | natural language processing |
The effect of diversity on group decision-making | arXiv:2402.01427v2 Announce Type: replace
Abstract: We explore different aspects of cognitive diversity and its effect on the success of group deliberation. To evaluate this, we use 500 dialogues from small, online groups discussing the Wason Card Selection task - the DeliData corpus. Leveraging the corpus, we perform quantitative analysis evaluating three different measures of cognitive diversity. First, we analyse the effect of group size as a proxy measure for diversity. Second, we evaluate the effect of the size of the initial idea pool. Finally, we look into the content of the discussion by analysing discussed solutions, discussion patterns, and how conversational probing can improve those characteristics. Despite the reputation of groups for compounding bias, we show that small groups can, through dialogue, overcome intuitive biases and improve individual decision-making. Across a large sample and different operationalisations, we consistently find that greater cognitive diversity is associated with more successful group deliberation. Code and data used for the analysis are available in the repository: https://github.com/gkaradzhov/cognitive-diversity-groups-cogsci24. | natural language processing |
Merging Facts, Crafting Fallacies: Evaluating the Contradictory Nature of Aggregated Factual Claims in Long-Form Generations | arXiv:2402.05629v3 Announce Type: replace
Abstract: Long-form generations from large language models (LLMs) contains a mix of factual and non-factual claims, making evaluating factuality difficult. To evaluate factual precision of long-form generations in a more fine-grained way, prior works propose to decompose long-form generations into multiple verifiable facts and verify those facts independently. The factuality of the generation is the proportion of verifiable facts among all the facts. Such methods assume that combining factual claims forms a factual paragraph. This paper shows that the assumption can be violated due to entity ambiguity. We show that LLMs can generate paragraphs that contain verifiable facts, but the facts are combined to form a non-factual paragraph due to entity ambiguity. We further reveal that existing factual precision metrics, including FActScore and citation recall, cannot properly evaluate the factuality of these non-factual paragraphs. To address this, we introduce an enhanced metric, D-FActScore, specifically designed for content with ambiguous entities. We evaluate the D-FActScores of people biographies generated with retrieval-augmented generation (RAG). We show that D-FActScore can better assess the factuality of paragraphs with entity ambiguity than FActScore. We also find that four widely used open-source LLMs tend to mix information of distinct entities to form non-factual paragraphs. | natural language processing |
Generation Meets Verification: Accelerating Large Language Model Inference with Smart Parallel Auto-Correct Decoding | arXiv:2402.11809v3 Announce Type: replace
Abstract: This research aims to accelerate the inference speed of large language models (LLMs) with billions of parameters. We propose \textbf{S}mart \textbf{P}arallel \textbf{A}uto-\textbf{C}orrect d\textbf{E}coding (SPACE), an innovative approach designed for achieving lossless acceleration of LLMs. By integrating semi-autoregressive inference and speculative decoding capabilities, SPACE uniquely enables autoregressive LLMs to parallelize token generation and verification. This is realized through a specialized semi-autoregressive supervised fine-tuning process that equips existing LLMs with the ability to simultaneously predict multiple tokens. Additionally, an auto-correct decoding algorithm facilitates the simultaneous generation and verification of token sequences within a single model invocation. Through extensive experiments on a range of LLMs, SPACE has demonstrated inference speedup ranging from 2.7x-4.0x on HumanEval-X while maintaining output quality. | natural language processing |
API-BLEND: A Comprehensive Corpora for Training and Benchmarking API LLMs | arXiv:2402.15491v2 Announce Type: replace
Abstract: There is a growing need for Large Language Models (LLMs) to effectively use tools and external Application Programming Interfaces (APIs) to plan and complete tasks. As such, there is tremendous interest in methods that can acquire sufficient quantities of train and test data that involve calls to tools / APIs. Two lines of research have emerged as the predominant strategies for addressing this challenge. The first has focused on synthetic data generation techniques, while the second has involved curating task-adjacent datasets which can be transformed into API / Tool-based tasks. In this paper, we focus on the task of identifying, curating, and transforming existing datasets and, in turn, introduce API-BLEND, a large corpora for training and systematic testing of tool-augmented LLMs. The datasets mimic real-world scenarios involving API-tasks such as API / tool detection, slot filling, and sequencing of the detected APIs. We demonstrate the utility of the API-BLEND dataset for both training and benchmarking purposes. | natural language processing |
HiGPT: Heterogeneous Graph Language Model | arXiv:2402.16024v2 Announce Type: replace
Abstract: Heterogeneous graph learning aims to capture complex relationships and diverse relational semantics among entities in a heterogeneous graph to obtain meaningful representations for nodes and edges. Recent advancements in heterogeneous graph neural networks (HGNNs) have achieved state-of-the-art performance by considering relation heterogeneity and using specialized message functions and aggregation rules. However, existing frameworks for heterogeneous graph learning have limitations in generalizing across diverse heterogeneous graph datasets. Most of these frameworks follow the "pre-train" and "fine-tune" paradigm on the same dataset, which restricts their capacity to adapt to new and unseen data. This raises the question: "Can we generalize heterogeneous graph models to be well-adapted to diverse downstream learning tasks with distribution shifts in both node token sets and relation type heterogeneity?'' To tackle those challenges, we propose HiGPT, a general large graph model with Heterogeneous graph instruction-tuning paradigm. Our framework enables learning from arbitrary heterogeneous graphs without the need for any fine-tuning process from downstream datasets. To handle distribution shifts in heterogeneity, we introduce an in-context heterogeneous graph tokenizer that captures semantic relationships in different heterogeneous graphs, facilitating model adaptation. We incorporate a large corpus of heterogeneity-aware graph instructions into our HiGPT, enabling the model to effectively comprehend complex relation heterogeneity and distinguish between various types of graph tokens. Furthermore, we introduce the Mixture-of-Thought (MoT) instruction augmentation paradigm to mitigate data scarcity by generating diverse and informative instructions. Through comprehensive evaluations, our proposed framework demonstrates exceptional performance in terms of generalization performance. | natural language processing |
M3-VRD: Multimodal Multi-task Multi-teacher Visually-Rich Form Document Understanding | arXiv:2402.17983v2 Announce Type: replace
Abstract: This paper presents a groundbreaking multimodal, multi-task, multi-teacher joint-grained knowledge distillation model for visually-rich form document understanding. The model is designed to leverage insights from both fine-grained and coarse-grained levels by facilitating a nuanced correlation between token and entity representations, addressing the complexities inherent in form documents. Additionally, we introduce new inter-grained and cross-grained loss functions to further refine diverse multi-teacher knowledge distillation transfer process, presenting distribution gaps and a harmonised understanding of form documents. Through a comprehensive evaluation across publicly available form document understanding datasets, our proposed model consistently outperforms existing baselines, showcasing its efficacy in handling the intricate structures and content of visually complex form documents. | computer vision |
UrbanGPT: Spatio-Temporal Large Language Models | arXiv:2403.00813v3 Announce Type: replace
Abstract: Spatio-temporal prediction aims to forecast and gain insights into the ever-changing dynamics of urban environments across both time and space. Its purpose is to anticipate future patterns, trends, and events in diverse facets of urban life, including transportation, population movement, and crime rates. Although numerous efforts have been dedicated to developing neural network techniques for accurate predictions on spatio-temporal data, it is important to note that many of these methods heavily depend on having sufficient labeled data to generate precise spatio-temporal representations. Unfortunately, the issue of data scarcity is pervasive in practical urban sensing scenarios. Consequently, it becomes necessary to build a spatio-temporal model with strong generalization capabilities across diverse spatio-temporal learning scenarios. Taking inspiration from the remarkable achievements of large language models (LLMs), our objective is to create a spatio-temporal LLM that can exhibit exceptional generalization capabilities across a wide range of downstream urban tasks. To achieve this objective, we present the UrbanGPT, which seamlessly integrates a spatio-temporal dependency encoder with the instruction-tuning paradigm. This integration enables LLMs to comprehend the complex inter-dependencies across time and space, facilitating more comprehensive and accurate predictions under data scarcity. To validate the effectiveness of our approach, we conduct extensive experiments on various public datasets, covering different spatio-temporal prediction tasks. The results consistently demonstrate that our UrbanGPT, with its carefully designed architecture, consistently outperforms state-of-the-art baselines. These findings highlight the potential of building large language models for spatio-temporal learning, particularly in zero-shot scenarios where labeled data is scarce. | natural language processing |
Detecting AI-Generated Sentences in Realistic Human-AI Collaborative Hybrid Texts: Challenges, Strategies, and Insights | arXiv:2403.03506v3 Announce Type: replace
Abstract: This study explores the challenge of sentence-level AI-generated text detection within human-AI collaborative hybrid texts. Existing studies of AI-generated text detection for hybrid texts often rely on synthetic datasets. These typically involve hybrid texts with a limited number of boundaries. We contend that studies of detecting AI-generated content within hybrid texts should cover different types of hybrid texts generated in realistic settings to better inform real-world applications. Therefore, our study utilizes the CoAuthor dataset, which includes diverse, realistic hybrid texts generated through the collaboration between human writers and an intelligent writing system in multi-turn interactions. We adopt a two-step, segmentation-based pipeline: (i) detect segments within a given hybrid text where each segment contains sentences of consistent authorship, and (ii) classify the authorship of each identified segment. Our empirical findings highlight (1) detecting AI-generated sentences in hybrid texts is overall a challenging task because (1.1) human writers' selecting and even editing AI-generated sentences based on personal preferences adds difficulty in identifying the authorship of segments; (1.2) the frequent change of authorship between neighboring sentences within the hybrid text creates difficulties for segment detectors in identifying authorship-consistent segments; (1.3) the short length of text segments within hybrid texts provides limited stylistic cues for reliable authorship determination; (2) before embarking on the detection process, it is beneficial to assess the average length of segments within the hybrid text. This assessment aids in deciding whether (2.1) to employ a text segmentation-based strategy for hybrid texts with longer segments, or (2.2) to adopt a direct sentence-by-sentence classification strategy for those with shorter segments. | natural language processing |
Detoxifying Large Language Models via Knowledge Editing | arXiv:2403.14472v4 Announce Type: replace
Abstract: This paper investigates using knowledge editing techniques to detoxify Large Language Models (LLMs). We construct a benchmark, SafeEdit, which covers nine unsafe categories with various powerful attack prompts and equips comprehensive metrics for systematic evaluation. We conduct experiments with several knowledge editing approaches, indicating that knowledge editing has the potential to detoxify LLMs with a limited impact on general performance efficiently. Then, we propose a simple yet effective baseline, dubbed Detoxifying with Intraoperative Neural Monitoring (DINM), to diminish the toxicity of LLMs within a few tuning steps via only one instance. We further provide an in-depth analysis of the internal mechanism for various detoxifying approaches, demonstrating that previous methods like SFT and DPO may merely suppress the activations of toxic parameters, while DINM mitigates the toxicity of the toxic parameters to a certain extent, making permanent adjustments. We hope that these insights could shed light on future work of developing detoxifying approaches and the underlying knowledge mechanisms of LLMs. Code and benchmark are available at https://github.com/zjunlp/EasyEdit. | natural language processing |
LLMs Are Few-Shot In-Context Low-Resource Language Learners | arXiv:2403.16512v3 Announce Type: replace
Abstract: In-context learning (ICL) empowers large language models (LLMs) to perform diverse tasks in underrepresented languages using only short in-context information, offering a crucial avenue for narrowing the gap between high-resource and low-resource languages. Nonetheless, there is only a handful of works explored ICL for low-resource languages with most of them focusing on relatively high-resource languages, such as French and Spanish. In this work, we extensively study ICL and its cross-lingual variation (X-ICL) on 25 low-resource and 7 relatively higher-resource languages. Our study not only assesses the effectiveness of ICL with LLMs in low-resource languages but also identifies the shortcomings of in-context label alignment, and introduces a more effective alternative: query alignment. Moreover, we provide valuable insights into various facets of ICL for low-resource languages. Our study concludes the significance of few-shot in-context information on enhancing the low-resource understanding quality of LLMs through semantically relevant information by closing the language gap in the target language and aligning the semantics between the targeted low-resource and the high-resource language that the model is proficient in. Our work highlights the importance of advancing ICL research, particularly for low-resource languages. | natural language processing |
ToXCL: A Unified Framework for Toxic Speech Detection and Explanation | arXiv:2403.16685v2 Announce Type: replace
Abstract: The proliferation of online toxic speech is a pertinent problem posing threats to demographic groups. While explicit toxic speech contains offensive lexical signals, implicit one consists of coded or indirect language. Therefore, it is crucial for models not only to detect implicit toxic speech but also to explain its toxicity. This draws a unique need for unified frameworks that can effectively detect and explain implicit toxic speech. Prior works mainly formulated the task of toxic speech detection and explanation as a text generation problem. Nonetheless, models trained using this strategy can be prone to suffer from the consequent error propagation problem. Moreover, our experiments reveal that the detection results of such models are much lower than those that focus only on the detection task. To bridge these gaps, we introduce ToXCL, a unified framework for the detection and explanation of implicit toxic speech. Our model consists of three modules: a (i) Target Group Generator to generate the targeted demographic group(s) of a given post; an (ii) Encoder-Decoder Model in which the encoder focuses on detecting implicit toxic speech and is boosted by a (iii) Teacher Classifier via knowledge distillation, and the decoder generates the necessary explanation. ToXCL achieves new state-of-the-art effectiveness, and outperforms baselines significantly. | natural language processing |
Towards a Framework for Evaluating Explanations in Automated Fact Verification | arXiv:2403.20322v2 Announce Type: replace
Abstract: As deep neural models in NLP become more complex, and as a consequence opaque, the necessity to interpret them becomes greater. A burgeoning interest has emerged in rationalizing explanations to provide short and coherent justifications for predictions. In this position paper, we advocate for a formal framework for key concepts and properties about rationalizing explanations to support their evaluation systematically. We also outline one such formal framework, tailored to rationalizing explanations of increasingly complex structures, from free-form explanations to deductive explanations, to argumentative explanations (with the richest structure). Focusing on the automated fact verification task, we provide illustrations of the use and usefulness of our formalization for evaluating explanations, tailored to their varying structures. | natural language processing |
An image speaks a thousand words, but can everyone listen? On image transcreation for cultural relevance | arXiv:2404.01247v2 Announce Type: replace
Abstract: Given the rise of multimedia content, human translators increasingly focus on culturally adapting not only words but also other modalities such as images to convey the same meaning. While several applications stand to benefit from this, machine translation systems remain confined to dealing with language in speech and text. In this work, we take a first step towards translating images to make them culturally relevant. First, we build three pipelines comprising state-of-the-art generative models to do the task. Next, we build a two-part evaluation dataset: i) concept: comprising 600 images that are cross-culturally coherent, focusing on a single concept per image, and ii) application: comprising 100 images curated from real-world applications. We conduct a multi-faceted human evaluation of translated images to assess for cultural relevance and meaning preservation. We find that as of today, image-editing models fail at this task, but can be improved by leveraging LLMs and retrievers in the loop. Best pipelines can only translate 5% of images for some countries in the easier concept dataset and no translation is successful for some countries in the application dataset, highlighting the challenging nature of the task. Our code and data is released here: https://github.com/simran-khanuja/image-transcreation. | natural language processing |
Conversational Disease Diagnosis via External Planner-Controlled Large Language Models | arXiv:2404.04292v5 Announce Type: replace
Abstract: The development of large language models (LLMs) has brought unprecedented possibilities for artificial intelligence (AI) based medical diagnosis. However, the application perspective of LLMs in real diagnostic scenarios is still unclear because they are not adept at collecting patient data proactively. This study presents a LLM-based diagnostic system that enhances planning capabilities by emulating doctors. Our system involves two external planners to handle planning tasks. The first planner employs a reinforcement learning approach to formulate disease screening questions and conduct initial diagnoses. The second planner uses LLMs to parse medical guidelines and conduct differential diagnoses. By utilizing real patient electronic medical record data, we constructed simulated dialogues between virtual patients and doctors and evaluated the diagnostic abilities of our system. We demonstrated that our system obtained impressive performance in both disease screening and differential diagnoses tasks. This research represents a step towards more seamlessly integrating AI into clinical settings, potentially enhancing the accuracy and accessibility of medical diagnostics. | natural language processing |
Incubating Text Classifiers Following User Instruction with Nothing but LLM | arXiv:2404.10877v2 Announce Type: replace
Abstract: In this paper, we aim to generate text classification data given arbitrary class definitions (i.e., user instruction), so one can train a small text classifier without any human annotation or raw corpus. Compared with pioneer attempts, our proposed Incubator is the first framework that can handle complicated and even mutually dependent classes (e.g., "TED Talk given by Educator" and "Other"). Specifically, Incubator is an LLM firstly tuned on the instruction-to-data mappings that we obtained from classification datasets and descriptions on HuggingFace together with in-context augmentation by GPT-4. We then refine Incubator by learning on the cluster centers of semantic textual embeddings to emphasize the uniformity and semantic diversity in generations. We compare Incubator on various classification tasks with strong baselines such as direct LLM-based inference and training data generation by prompt engineering. Experiments show Incubator is able to (1) perform well on traditional benchmarks, (2) take label dependency and user preference into consideration, and (3) enable logical text mining by incubating multiple classifiers. | natural language processing |
CofiPara: A Coarse-to-fine Paradigm for Multimodal Sarcasm Target Identification with Large Multimodal Models | arXiv:2405.00390v2 Announce Type: replace
Abstract: Social media abounds with multimodal sarcasm, and identifying sarcasm targets is particularly challenging due to the implicit incongruity not directly evident in the text and image modalities. Current methods for Multimodal Sarcasm Target Identification (MSTI) predominantly focus on superficial indicators in an end-to-end manner, overlooking the nuanced understanding of multimodal sarcasm conveyed through both the text and image. This paper proposes a versatile MSTI framework with a coarse-to-fine paradigm, by augmenting sarcasm explainability with reasoning and pre-training knowledge. Inspired by the powerful capacity of Large Multimodal Models (LMMs) on multimodal reasoning, we first engage LMMs to generate competing rationales for coarser-grained pre-training of a small language model on multimodal sarcasm detection. We then propose fine-tuning the model for finer-grained sarcasm target identification. Our framework is thus empowered to adeptly unveil the intricate targets within multimodal sarcasm and mitigate the negative impact posed by potential noise inherently in LMMs. Experimental results demonstrate that our model far outperforms state-of-the-art MSTI methods, and markedly exhibits explainability in deciphering sarcasm as well. | natural language processing |
LLM Discussion: Enhancing the Creativity of Large Language Models via Discussion Framework and Role-Play | arXiv:2405.06373v3 Announce Type: replace
Abstract: Large language models (LLMs) have shown exceptional proficiency in natural language processing but often fall short of generating creative and original responses to open-ended questions. To enhance LLM creativity, our key insight is to emulate the human process of inducing collective creativity through engaging discussions with participants from diverse backgrounds and perspectives. To this end, we propose LLM Discussion, a three-phase discussion framework that facilitates vigorous and diverging idea exchanges and ensures convergence to creative answers. Moreover, we adopt a role-playing technique by assigning distinct roles to LLMs to combat the homogeneity of LLMs. We evaluate the efficacy of the proposed framework with the Alternative Uses Test, Similarities Test, Instances Test, and Scientific Creativity Test through both LLM evaluation and human study. Our proposed framework outperforms single-LLM approaches and existing multi-LLM frameworks across various creativity metrics. | natural language processing |
Improving Instruction Following in Language Models through Proxy-Based Uncertainty Estimation | arXiv:2405.06424v2 Announce Type: replace
Abstract: Assessing response quality to instructions in language models is vital but challenging due to the complexity of human language across different contexts. This complexity often results in ambiguous or inconsistent interpretations, making accurate assessment difficult. To address this issue, we propose a novel Uncertainty-aware Reward Model (URM) that introduces a robust uncertainty estimation for the quality of paired responses based on Bayesian approximation. Trained with preference datasets, our uncertainty-enabled proxy not only scores rewards for responses but also evaluates their inherent uncertainty. Empirical results demonstrate significant benefits of incorporating the proposed proxy into language model training. Our method boosts the instruction following capability of language models by refining data curation for training and improving policy optimization objectives, thereby surpassing existing methods by a large margin on benchmarks such as Vicuna and MT-bench. These findings highlight that our proposed approach substantially advances language model training and paves a new way of harnessing uncertainty within language models. | natural language processing |
PolygloToxicityPrompts: Multilingual Evaluation of Neural Toxic Degeneration in Large Language Models | arXiv:2405.09373v2 Announce Type: replace
Abstract: Recent advances in large language models (LLMs) have led to their extensive global deployment, and ensuring their safety calls for comprehensive and multilingual toxicity evaluations. However, existing toxicity benchmarks are overwhelmingly focused on English, posing serious risks to deploying LLMs in other languages. We address this by introducing PolygloToxicityPrompts (PTP), the first large-scale multilingual toxicity evaluation benchmark of 425K naturally occurring prompts spanning 17 languages. We overcome the scarcity of naturally occurring toxicity in web-text and ensure coverage across languages with varying resources by automatically scraping over 100M web-text documents. Using PTP, we investigate research questions to study the impact of model size, prompt language, and instruction and preference-tuning methods on toxicity by benchmarking over 60 LLMs. Notably, we find that toxicity increases as language resources decrease or model size increases. Although instruction- and preference-tuning reduce toxicity, the choice of preference-tuning method does not have any significant impact. Our findings shed light on crucial shortcomings of LLM safeguarding and highlight areas for future research. | natural language processing |
CPsyExam: A Chinese Benchmark for Evaluating Psychology using Examinations | arXiv:2405.10212v2 Announce Type: replace
Abstract: In this paper, we introduce a novel psychological benchmark, CPsyExam, constructed from questions sourced from Chinese language examinations. CPsyExam is designed to prioritize psychological knowledge and case analysis separately, recognizing the significance of applying psychological knowledge to real-world scenarios. From the pool of 22k questions, we utilize 4k to create the benchmark that offers balanced coverage of subjects and incorporates a diverse range of case analysis techniques.Furthermore, we evaluate a range of existing large language models~(LLMs), spanning from open-sourced to API-based models. Our experiments and analysis demonstrate that CPsyExam serves as an effective benchmark for enhancing the understanding of psychology within LLMs and enables the comparison of LLMs across various granularities. | natural language processing |
AmazUtah_NLP at SemEval-2024 Task 9: A MultiChoice Question Answering System for Commonsense Defying Reasoning | arXiv:2405.10385v2 Announce Type: replace
Abstract: The SemEval 2024 BRAINTEASER task represents a pioneering venture in Natural Language Processing (NLP) by focusing on lateral thinking, a dimension of cognitive reasoning that is often overlooked in traditional linguistic analyses. This challenge comprises of Sentence Puzzle and Word Puzzle subtasks and aims to test language models' capacity for divergent thinking.
In this paper, we present our approach to the BRAINTEASER task. We employ a holistic strategy by leveraging cutting-edge pre-trained models in multiple choice architecture, and diversify the training data with Sentence and Word Puzzle datasets. To gain further improvement, we fine-tuned the model with synthetic humor or jokes dataset and the RiddleSense dataset which helped augmenting the model's lateral thinking abilities. Empirical results show that our approach achieve 92.5% accuracy in Sentence Puzzle subtask and 80.2% accuracy in Word Puzzle subtask. | natural language processing |
Language Models can Exploit Cross-Task In-context Learning for Data-Scarce Novel Tasks | arXiv:2405.10548v2 Announce Type: replace
Abstract: Large Language Models (LLMs) have transformed NLP with their remarkable In-context Learning (ICL) capabilities. Automated assistants based on LLMs are gaining popularity; however, adapting them to novel tasks is still challenging. While colossal models excel in zero-shot performance, their computational demands limit widespread use, and smaller language models struggle without context. This paper investigates whether LLMs can generalize from labeled examples of predefined tasks to novel tasks. Drawing inspiration from biological neurons and the mechanistic interpretation of the Transformer architecture, we explore the potential for information sharing across tasks. We design a cross-task prompting setup with three LLMs and show that LLMs achieve significant performance improvements despite no examples from the target task in the context. Cross-task prompting leads to a remarkable performance boost of 107% for LLaMA-2 7B, 18.6% for LLaMA-2 13B, and 3.2% for GPT 3.5 on average over zero-shot prompting, and performs comparable to standard in-context learning. The effectiveness of generating pseudo-labels for in-task examples is demonstrated, and our analyses reveal a strong correlation between the effect of cross-task examples and model activation similarities in source and target input tokens. This paper offers a first-of-its-kind exploration of LLMs' ability to solve novel tasks based on contextual signals from different task examples. | natural language processing |
SPOR: A Comprehensive and Practical Evaluation Method for Compositional Generalization in Data-to-Text Generation | arXiv:2405.10650v2 Announce Type: replace
Abstract: Compositional generalization is an important ability of language models and has many different manifestations. For data-to-text generation, previous research on this ability is limited to a single manifestation called Systematicity and lacks consideration of large language models (LLMs), which cannot fully cover practical application scenarios. In this work, we propose SPOR, a comprehensive and practical evaluation method for compositional generalization in data-to-text generation. SPOR includes four aspects of manifestations (Systematicity, Productivity, Order invariance, and Rule learnability) and allows high-quality evaluation without additional manual annotations based on existing datasets. We demonstrate SPOR on two different datasets and evaluate some existing language models including LLMs. We find that the models are deficient in various aspects of the evaluation and need further improvement. Our work shows the necessity for comprehensive research on different manifestations of compositional generalization in data-to-text generation and provides a framework for evaluation. | natural language processing |
Realistic Evaluation of Toxicity in Large Language Models | arXiv:2405.10659v2 Announce Type: replace
Abstract: Large language models (LLMs) have become integral to our professional workflows and daily lives. Nevertheless, these machine companions of ours have a critical flaw: the huge amount of data which endows them with vast and diverse knowledge, also exposes them to the inevitable toxicity and bias. While most LLMs incorporate defense mechanisms to prevent the generation of harmful content, these safeguards can be easily bypassed with minimal prompt engineering. In this paper, we introduce the new Thoroughly Engineered Toxicity (TET) dataset, comprising manually crafted prompts designed to nullify the protective layers of such models. Through extensive evaluations, we demonstrate the pivotal role of TET in providing a rigorous benchmark for evaluation of toxicity awareness in several popular LLMs: it highlights the toxicity in the LLMs that might remain hidden when using normal prompts, thus revealing subtler issues in their behavior. | natural language processing |
Statler: State-Maintaining Language Models for Embodied Reasoning | arXiv:2306.17840v4 Announce Type: replace-cross
Abstract: There has been a significant research interest in employing large language models to empower intelligent robots with complex reasoning. Existing work focuses on harnessing their abilities to reason about the histories of their actions and observations. In this paper, we explore a new dimension in which large language models may benefit robotics planning. In particular, we propose Statler, a framework in which large language models are prompted to maintain an estimate of the world state, which are often unobservable, and track its transition as new actions are taken. Our framework then conditions each action on the estimate of the current world state. Despite being conceptually simple, our Statler framework significantly outperforms strong competing methods (e.g., Code-as-Policies) on several robot planning tasks. Additionally, it has the potential advantage of scaling up to more challenging long-horizon planning tasks. | natural language processing |
A Hybrid Machine Learning Model for Classifying Gene Mutations in Cancer using LSTM, BiLSTM, CNN, GRU, and GloVe | arXiv:2307.14361v3 Announce Type: replace-cross
Abstract: In our study, we introduce a novel hybrid ensemble model that synergistically combines LSTM, BiLSTM, CNN, GRU, and GloVe embeddings for the classification of gene mutations in cancer. This model was rigorously tested using Kaggle's Personalized Medicine: Redefining Cancer Treatment dataset, demonstrating exceptional performance across all evaluation metrics. Notably, our approach achieved a training accuracy of 80.6%, precision of 81.6%, recall of 80.6%, and an F1 score of 83.1%, alongside a significantly reduced Mean Squared Error (MSE) of 2.596. These results surpass those of advanced transformer models and their ensembles, showcasing our model's superior capability in handling the complexities of gene mutation classification. The accuracy and efficiency of gene mutation classification are paramount in the era of precision medicine, where tailored treatment plans based on individual genetic profiles can dramatically improve patient outcomes and save lives. Our model's remarkable performance highlights its potential in enhancing the precision of cancer diagnoses and treatments, thereby contributing significantly to the advancement of personalized healthcare. | computer vision |
A Semantic Invariant Robust Watermark for Large Language Models | arXiv:2310.06356v3 Announce Type: replace-cross
Abstract: Watermark algorithms for large language models (LLMs) have achieved extremely high accuracy in detecting text generated by LLMs. Such algorithms typically involve adding extra watermark logits to the LLM's logits at each generation step. However, prior algorithms face a trade-off between attack robustness and security robustness. This is because the watermark logits for a token are determined by a certain number of preceding tokens; a small number leads to low security robustness, while a large number results in insufficient attack robustness. In this work, we propose a semantic invariant watermarking method for LLMs that provides both attack robustness and security robustness. The watermark logits in our work are determined by the semantics of all preceding tokens. Specifically, we utilize another embedding LLM to generate semantic embeddings for all preceding tokens, and then these semantic embeddings are transformed into the watermark logits through our trained watermark model. Subsequent analyses and experiments demonstrated the attack robustness of our method in semantically invariant settings: synonym substitution and text paraphrasing settings. Finally, we also show that our watermark possesses adequate security robustness. Our code and data are available at \href{https://github.com/THU-BPM/Robust_Watermark}{https://github.com/THU-BPM/Robust\_Watermark}. Additionally, our algorithm could also be accessed through MarkLLM \citep{pan2024markllm} \footnote{https://github.com/THU-BPM/MarkLLM}. | natural language processing |
A Framework for Inference Inspired by Human Memory Mechanisms | arXiv:2310.09297v2 Announce Type: replace-cross
Abstract: How humans and machines make sense of current inputs for relation reasoning and question-answering while putting the perceived information into context of our past memories, has been a challenging conundrum in cognitive science and artificial intelligence. Inspired by human brain's memory system and cognitive architectures, we propose a PMI framework that consists of perception, memory and inference components. Notably, the memory module comprises working and long-term memory, with the latter endowed with a higher-order structure to retain extensive and complex relational knowledge and experience. Through a differentiable competitive write access, current perceptions update working memory, which is later merged with long-term memory via outer product associations, reducing information conflicts and averting memory overflow. In the inference module, relevant information is retrieved from two separate memory origins and associatively integrated to attain a more comprehensive and precise interpretation of current perceptions. We exploratively apply our PMI to improve prevailing Transformers and CNN models on question-answering tasks like bAbI-20k and Sort-of-CLEVR datasets, as well as detecting equilateral triangles, language modeling and image classification tasks, and in each case, our PMI enhancements consistently outshine their original counterparts significantly. Visualization analyses reveal that relational memory consolidation, along with the interaction and integration of information from diverse memory sources, substantially contributes to the model effectiveness on inference tasks. | natural language processing |
Universal Prompt Optimizer for Safe Text-to-Image Generation | arXiv:2402.10882v2 Announce Type: replace-cross
Abstract: Text-to-Image (T2I) models have shown great performance in generating images based on textual prompts. However, these models are vulnerable to unsafe input to generate unsafe content like sexual, harassment and illegal-activity images. Existing studies based on image checker, model fine-tuning and embedding blocking are impractical in real-world applications. Hence, we propose the first universal prompt optimizer for safe T2I (POSI) generation in black-box scenario. We first construct a dataset consisting of toxic-clean prompt pairs by GPT-3.5 Turbo. To guide the optimizer to have the ability of converting toxic prompt to clean prompt while preserving semantic information, we design a novel reward function measuring toxicity and text alignment of generated images and train the optimizer through Proximal Policy Optimization. Experiments show that our approach can effectively reduce the likelihood of various T2I models in generating inappropriate images, with no significant impact on text alignment. It is also flexible to be combined with methods to achieve better performance. Our code is available at https://github.com/wzongyu/POSI. | computer vision |
RALL-E: Robust Codec Language Modeling with Chain-of-Thought Prompting for Text-to-Speech Synthesis | arXiv:2404.03204v3 Announce Type: replace-cross
Abstract: We present RALL-E, a robust language modeling method for text-to-speech (TTS) synthesis. While previous work based on large language models (LLMs) shows impressive performance on zero-shot TTS, such methods often suffer from poor robustness, such as unstable prosody (weird pitch and rhythm/duration) and a high word error rate (WER), due to the autoregressive prediction style of language models. The core idea behind RALL-E is chain-of-thought (CoT) prompting, which decomposes the task into simpler steps to enhance the robustness of LLM-based TTS. To accomplish this idea, RALL-E first predicts prosody features (pitch and duration) of the input text and uses them as intermediate conditions to predict speech tokens in a CoT style. Second, RALL-E utilizes the predicted duration prompt to guide the computing of self-attention weights in Transformer to enforce the model to focus on the corresponding phonemes and prosody features when predicting speech tokens. Results of comprehensive objective and subjective evaluations demonstrate that, compared to a powerful baseline method VALL-E, RALL-E significantly improves the WER of zero-shot TTS from $5.6\%$ (without reranking) and $1.7\%$ (with reranking) to $2.5\%$ and $1.0\%$, respectively. Furthermore, we demonstrate that RALL-E correctly synthesizes sentences that are hard for VALL-E and reduces the error rate from $68\%$ to $4\%$. | natural language processing |
Language Model Prompt Selection via Simulation Optimization | arXiv:2404.08164v2 Announce Type: replace-cross
Abstract: With the advancement in generative language models, the selection of prompts has gained significant attention in recent years. A prompt is an instruction or description provided by the user, serving as a guide for the generative language model in content generation. Despite existing methods for prompt selection that are based on human labor, we consider facilitating this selection through simulation optimization, aiming to maximize a pre-defined score for the selected prompt. Specifically, we propose a two-stage framework. In the first stage, we determine a feasible set of prompts in sufficient numbers, where each prompt is represented by a moderate-dimensional vector. In the subsequent stage for evaluation and selection, we construct a surrogate model of the score regarding the moderate-dimensional vectors that represent the prompts. We propose sequentially selecting the prompt for evaluation based on this constructed surrogate model. We prove the consistency of the sequential evaluation procedure in our framework. We also conduct numerical experiments to demonstrate the efficacy of our proposed framework, providing practical instructions for implementation. | natural language processing |
Special Characters Attack: Toward Scalable Training Data Extraction From Large Language Models | arXiv:2405.05990v2 Announce Type: replace-cross
Abstract: Large language models (LLMs) have achieved remarkable performance on a wide range of tasks. However, recent studies have shown that LLMs can memorize training data and simple repeated tokens can trick the model to leak the data. In this paper, we take a step further and show that certain special characters or their combinations with English letters are stronger memory triggers, leading to more severe data leakage. The intuition is that, since LLMs are trained with massive data that contains a substantial amount of special characters (e.g. structural symbols {, } of JSON files, and @, # in emails and online posts), the model may memorize the co-occurrence between these special characters and the raw texts. This motivates us to propose a simple but effective Special Characters Attack (SCA) to induce training data leakage. Our experiments verify the high effectiveness of SCA against state-of-the-art LLMs: they can leak diverse training data, such as code corpus, web pages, and personally identifiable information, and sometimes generate non-stop outputs as a byproduct. We further show that the composition of the training data corpus can be revealed by inspecting the leaked data -- one crucial piece of information for pre-training high-performance LLMs. Our work can help understand the sensitivity of LLMs to special characters and identify potential areas for improvement. | natural language processing |
LangCell: Language-Cell Pre-training for Cell Identity Understanding | arXiv:2405.06708v2 Announce Type: replace-cross
Abstract: Cell identity encompasses various semantic aspects of a cell, including cell type, pathway information, disease information, and more, which are essential for biologists to gain insights into its biological characteristics. Understanding cell identity from the transcriptomic data, such as annotating cell types, have become an important task in bioinformatics. As these semantic aspects are determined by human experts, it is impossible for AI models to effectively carry out cell identity understanding tasks without the supervision signals provided by single-cell and label pairs. The single-cell pre-trained language models (PLMs) currently used for this task are trained only on a single modality, transcriptomics data, lack an understanding of cell identity knowledge. As a result, they have to be fine-tuned for downstream tasks and struggle when lacking labeled data with the desired semantic labels. To address this issue, we propose an innovative solution by constructing a unified representation of single-cell data and natural language during the pre-training phase, allowing the model to directly incorporate insights related to cell identity. More specifically, we introduce LangCell, the first Language-Cell pre-training framework. LangCell utilizes texts enriched with cell identity information to gain a profound comprehension of cross-modal knowledge. Results from experiments conducted on different benchmarks show that LangCell is the only single-cell PLM that can work effectively in zero-shot cell identity understanding scenarios, and also significantly outperforms existing models in few-shot and fine-tuning cell identity understanding scenarios. | natural language processing |
Decoding Geometric Properties in Non-Random Data from First Information-Theoretic Principles | arXiv:2405.07803v2 Announce Type: replace-cross
Abstract: Based on the principles of information theory, measure theory, and theoretical computer science, we introduce a univariate signal deconvolution method with a wide range of applications to coding theory, particularly in zero-knowledge one-way communication channels, such as in deciphering messages from unknown generating sources about which no prior knowledge is available and to which no return message can be sent. Our multidimensional space reconstruction method from an arbitrary received signal is proven to be agnostic vis-a-vis the encoding-decoding scheme, computation model, programming language, formal theory, the computable (or semi-computable) method of approximation to algorithmic complexity, and any arbitrarily chosen (computable) probability measure of the events. The method derives from the principles of an approach to Artificial General Intelligence capable of building a general-purpose model of models independent of any arbitrarily assumed prior probability distribution. We argue that this optimal and universal method of decoding non-random data has applications to signal processing, causal deconvolution, topological and geometric properties encoding, cryptography, and bio- and technosignature detection. | natural language processing |
Unveiling Hallucination in Text, Image, Video, and Audio Foundation Models: A Comprehensive Survey | arXiv:2405.09589v2 Announce Type: replace-cross
Abstract: The rapid advancement of foundation models (FMs) across language, image, audio, and video domains has shown remarkable capabilities in diverse tasks. However, the proliferation of FMs brings forth a critical challenge: the potential to generate hallucinated outputs, particularly in high-stakes applications. The tendency of foundation models to produce hallucinated content arguably represents the biggest hindrance to their widespread adoption in real-world scenarios, especially in domains where reliability and accuracy are paramount. This survey paper presents a comprehensive overview of recent developments that aim to identify and mitigate the problem of hallucination in FMs, spanning text, image, video, and audio modalities. By synthesizing recent advancements in detecting and mitigating hallucination across various modalities, the paper aims to provide valuable insights for researchers, developers, and practitioners. Essentially, it establishes a clear framework encompassing definition, taxonomy, and detection strategies for addressing hallucination in multimodal foundation models, laying the foundation for future research in this pivotal area. | natural language processing |
Application of Tensorized Neural Networks for Cloud Classification | arXiv:2405.10946v1 Announce Type: new
Abstract: Convolutional neural networks (CNNs) have gained widespread usage across various fields such as weather forecasting, computer vision, autonomous driving, and medical image analysis due to its exceptional ability to extract spatial information, share parameters, and learn local features. However, the practical implementation and commercialization of CNNs in these domains are hindered by challenges related to model sizes, overfitting, and computational time. To address these limitations, our study proposes a groundbreaking approach that involves tensorizing the dense layers in the CNN to reduce model size and computational time. Additionally, we incorporate attention layers into the CNN and train it using Contrastive self-supervised learning to effectively classify cloud information, which is crucial for accurate weather forecasting. We elucidate the key characteristics of tensorized neural network (TNN), including the data compression rate, accuracy, and computational speed. The results indicate how TNN change their properties under the batch size setting. | natural language processing |
Depth-aware Panoptic Segmentation | arXiv:2405.10947v1 Announce Type: new
Abstract: Panoptic segmentation unifies semantic and instance segmentation and thus delivers a semantic class label and, for so-called thing classes, also an instance label per pixel. The differentiation of distinct objects of the same class with a similar appearance is particularly challenging and frequently causes such objects to be incorrectly assigned to a single instance. In the present work, we demonstrate that information on the 3D geometry of the observed scene can be used to mitigate this issue: We present a novel CNN-based method for panoptic segmentation which processes RGB images and depth maps given as input in separate network branches and fuses the resulting feature maps in a late fusion manner. Moreover, we propose a new depth-aware dice loss term which penalises the assignment of pixels to the same thing instance based on the difference between their associated distances to the camera. Experiments carried out on the Cityscapes dataset show that the proposed method reduces the number of objects that are erroneously merged into one thing instance and outperforms the method used as basis by 2.2% in terms of panoptic quality. | computer vision |
Surgical-LVLM: Learning to Adapt Large Vision-Language Model for Grounded Visual Question Answering in Robotic Surgery | arXiv:2405.10948v1 Announce Type: new
Abstract: Recent advancements in Surgical Visual Question Answering (Surgical-VQA) and related region grounding have shown great promise for robotic and medical applications, addressing the critical need for automated methods in personalized surgical mentorship. However, existing models primarily provide simple structured answers and struggle with complex scenarios due to their limited capability in recognizing long-range dependencies and aligning multimodal information. In this paper, we introduce Surgical-LVLM, a novel personalized large vision-language model tailored for complex surgical scenarios. Leveraging the pre-trained large vision-language model and specialized Visual Perception LoRA (VP-LoRA) blocks, our model excels in understanding complex visual-language tasks within surgical contexts. In addressing the visual grounding task, we propose the Token-Interaction (TIT) module, which strengthens the interaction between the grounding module and the language responses of the Large Visual Language Model (LVLM) after projecting them into the latent space. We demonstrate the effectiveness of Surgical-LVLM on several benchmarks, including EndoVis-17-VQLA, EndoVis-18-VQLA, and a newly introduced EndoVis Conversations dataset, which sets new performance standards. Our work contributes to advancing the field of automated surgical mentorship by providing a context-aware solution. | natural language processing |
Global License Plate Dataset | arXiv:2405.10949v1 Announce Type: new
Abstract: In the pursuit of advancing the state-of-the-art (SOTA) in road safety, traffic monitoring, surveillance, and logistics automation, we introduce the Global License Plate Dataset (GLPD). The dataset consists of over 5 million images, including diverse samples captured from 74 countries with meticulous annotations, including license plate characters, license plate segmentation masks, license plate corner vertices, as well as vehicle make, colour, and model. We also include annotated data on more classes, such as pedestrians, vehicles, roads, etc. We include a statistical analysis of the dataset, and provide baseline efficient and accurate models. The GLPD aims to be the primary benchmark dataset for model development and finetuning for license plate recognition. | computer vision |
Block Selective Reprogramming for On-device Training of Vision Transformers | arXiv:2405.10951v1 Announce Type: new
Abstract: The ubiquity of vision transformers (ViTs) for various edge applications, including personalized learning, has created the demand for on-device fine-tuning. However, training with the limited memory and computation power of edge devices remains a significant challenge. In particular, the memory required for training is much higher than that needed for inference, primarily due to the need to store activations across all layers in order to compute the gradients needed for weight updates. Previous works have explored reducing this memory requirement via frozen-weight training as well storing the activations in a compressed format. However, these methods are deemed inefficient due to their inability to provide training or inference speedup. In this paper, we first investigate the limitations of existing on-device training methods aimed at reducing memory and compute requirements. We then present block selective reprogramming (BSR) in which we fine-tune only a fraction of total blocks of a pre-trained model and selectively drop tokens based on self-attention scores of the frozen layers. To show the efficacy of BSR, we present extensive evaluations on ViT-B and DeiT-S with five different datasets. Compared to the existing alternatives, our approach simultaneously reduces training memory by up to 1.4x and compute cost by up to 2x while maintaining similar accuracy. We also showcase results for Mixture-of-Expert (MoE) models, demonstrating the effectiveness of our approach in multitask learning scenarios. | computer vision |
VICAN: Very Efficient Calibration Algorithm for Large Camera Networks | arXiv:2405.10952v1 Announce Type: new
Abstract: The precise estimation of camera poses within large camera networks is a foundational problem in computer vision and robotics, with broad applications spanning autonomous navigation, surveillance, and augmented reality. In this paper, we introduce a novel methodology that extends state-of-the-art Pose Graph Optimization (PGO) techniques. Departing from the conventional PGO paradigm, which primarily relies on camera-camera edges, our approach centers on the introduction of a dynamic element - any rigid object free to move in the scene - whose pose can be reliably inferred from a single image. Specifically, we consider the bipartite graph encompassing cameras, object poses evolving dynamically, and camera-object relative transformations at each time step. This shift not only offers a solution to the challenges encountered in directly estimating relative poses between cameras, particularly in adverse environments, but also leverages the inclusion of numerous object poses to ameliorate and integrate errors, resulting in accurate camera pose estimates. Though our framework retains compatibility with traditional PGO solvers, its efficacy benefits from a custom-tailored optimization scheme. To this end, we introduce an iterative primal-dual algorithm, capable of handling large graphs. Empirical benchmarks, conducted on a new dataset of simulated indoor environments, substantiate the efficacy and efficiency of our approach. | computer vision |
Multimodal CLIP Inference for Meta-Few-Shot Image Classification | arXiv:2405.10954v1 Announce Type: new
Abstract: In recent literature, few-shot classification has predominantly been defined by the N-way k-shot meta-learning problem. Models designed for this purpose are usually trained to excel on standard benchmarks following a restricted setup, excluding the use of external data. Given the recent advancements in large language and vision models, a question naturally arises: can these models directly perform well on meta-few-shot learning benchmarks? Multimodal foundation models like CLIP, which learn a joint (image, text) embedding, are of particular interest. Indeed, multimodal training has proven to enhance model robustness, especially regarding ambiguities, a limitation frequently observed in the few-shot setup. This study demonstrates that combining modalities from CLIP's text and image encoders outperforms state-of-the-art meta-few-shot learners on widely adopted benchmarks, all without additional training. Our results confirm the potential and robustness of multimodal foundation models like CLIP and serve as a baseline for existing and future approaches leveraging such models. | computer vision |
Photorealistic 3D Urban Scene Reconstruction and Point Cloud Extraction using Google Earth Imagery and Gaussian Splatting | arXiv:2405.11021v1 Announce Type: new
Abstract: 3D urban scene reconstruction and modelling is a crucial research area in remote sensing with numerous applications in academia, commerce, industry, and administration. Recent advancements in view synthesis models have facilitated photorealistic 3D reconstruction solely from 2D images. Leveraging Google Earth imagery, we construct a 3D Gaussian Splatting model of the Waterloo region centered on the University of Waterloo and are able to achieve view-synthesis results far exceeding previous 3D view-synthesis results based on neural radiance fields which we demonstrate in our benchmark. Additionally, we retrieved the 3D geometry of the scene using the 3D point cloud extracted from the 3D Gaussian Splatting model which we benchmarked against our Multi- View-Stereo dense reconstruction of the scene, thereby reconstructing both the 3D geometry and photorealistic lighting of the large-scale urban scene through 3D Gaussian Splatting | computer vision |
Bayesian Learning-driven Prototypical Contrastive Loss for Class-Incremental Learning | arXiv:2405.11067v1 Announce Type: new
Abstract: The primary objective of methods in continual learning is to learn tasks in a sequential manner over time from a stream of data, while mitigating the detrimental phenomenon of catastrophic forgetting. In this paper, we focus on learning an optimal representation between previous class prototypes and newly encountered ones. We propose a prototypical network with a Bayesian learning-driven contrastive loss (BLCL) tailored specifically for class-incremental learning scenarios. Therefore, we introduce a contrastive loss that incorporates new classes into the latent representation by reducing the intra-class distance and increasing the inter-class distance. Our approach dynamically adapts the balance between the cross-entropy and contrastive loss functions with a Bayesian learning technique. Empirical evaluations conducted on both the CIFAR-10 dataset for image classification and images of a GNSS-based dataset for interference classification validate the efficacy of our method, showcasing its superiority over existing state-of-the-art approaches. | natural language processing |
Enhancing Understanding Through Wildlife Re-Identification | arXiv:2405.11112v1 Announce Type: new
Abstract: We explore the field of wildlife re-identification by implementing an MLP from scratch using NumPy, A DCNN using Keras, and a binary classifier with LightGBM for the purpose of learning for an assignment. Analyzing the performance of multiple models on multiple datasets. We attempt to replicate prior research in metric learning for wildlife re-identification. Firstly, we find that the usage of MLPs trained for classification, then removing the output layer and using the second last layer as an embedding was not a successful strategy for similar learning; it seems like losses designed for embeddings such as triplet loss are required. The DCNNS performed well on some datasets but poorly on others, which did not align with findings in previous literature. The LightGBM classifier overfitted too heavily and was not significantly better than a constant model when trained and evaluated on all pairs using accuracy as a metric. The technical implementations used seem to match standards according to comparisons with documentation examples and good results on certain datasets. However, there is still more to explore in regards to being able to fully recreate past literature. | natural language processing |
Flexible Motion In-betweening with Diffusion Models | arXiv:2405.11126v1 Announce Type: new
Abstract: Motion in-betweening, a fundamental task in character animation, consists of generating motion sequences that plausibly interpolate user-provided keyframe constraints. It has long been recognized as a labor-intensive and challenging process. We investigate the potential of diffusion models in generating diverse human motions guided by keyframes. Unlike previous inbetweening methods, we propose a simple unified model capable of generating precise and diverse motions that conform to a flexible range of user-specified spatial constraints, as well as text conditioning. To this end, we propose Conditional Motion Diffusion In-betweening (CondMDI) which allows for arbitrary dense-or-sparse keyframe placement and partial keyframe constraints while generating high-quality motions that are diverse and coherent with the given keyframes. We evaluate the performance of CondMDI on the text-conditioned HumanML3D dataset and demonstrate the versatility and efficacy of diffusion models for keyframe in-betweening. We further explore the use of guidance and imputation-based approaches for inference-time keyframing and compare CondMDI against these methods. | computer vision |
MotionGS : Compact Gaussian Splatting SLAM by Motion Filter | arXiv:2405.11129v1 Announce Type: new
Abstract: With their high-fidelity scene representation capability, the attention of SLAM field is deeply attracted by the Neural Radiation Field (NeRF) and 3D Gaussian Splatting (3DGS). Recently, there has been a Surge in NeRF-based SLAM, while 3DGS-based SLAM is sparse. A novel 3DGS-based SLAM approach with a fusion of deep visual feature, dual keyframe selection and 3DGS is presented in this paper. Compared with the existing methods, the proposed selectively tracking is achieved by feature extraction and motion filter on each frame. The joint optimization of pose and 3D Gaussian runs through the entire mapping process. Additionally, the coarse-to-fine pose estimation and compact Gaussian scene representation are implemented by dual keyfeature selection and novel loss functions. Experimental results demonstrate that the proposed algorithm not only outperforms the existing methods in tracking and mapping, but also has less memory usage. | computer vision |
Detecting Multimodal Situations with Insufficient Context and Abstaining from Baseless Predictions | arXiv:2405.11145v1 Announce Type: new
Abstract: Despite the widespread adoption of Vision-Language Understanding (VLU) benchmarks such as VQA v2, OKVQA, A-OKVQA, GQA, VCR, SWAG, and VisualCOMET, our analysis reveals a pervasive issue affecting their integrity: these benchmarks contain samples where answers rely on assumptions unsupported by the provided context. Training models on such data foster biased learning and hallucinations as models tend to make similar unwarranted assumptions. To address this issue, we collect contextual data for each sample whenever available and train a context selection module to facilitate evidence-based model predictions. Strong improvements across multiple benchmarks demonstrate the effectiveness of our approach. Further, we develop a general-purpose Context-AwaRe Abstention (CARA) detector to identify samples lacking sufficient context and enhance model accuracy by abstaining from responding if the required context is absent. CARA exhibits generalization to new benchmarks it wasn't trained on, underscoring its utility for future VLU benchmarks in detecting or cleaning samples with inadequate context. Finally, we curate a Context Ambiguity and Sufficiency Evaluation (CASE) set to benchmark the performance of insufficient context detectors. Overall, our work represents a significant advancement in ensuring that vision-language models generate trustworthy and evidence-based outputs in complex real-world scenarios. | computer vision |
Multi-scale Information Sharing and Selection Network with Boundary Attention for Polyp Segmentation | arXiv:2405.11151v1 Announce Type: new
Abstract: Polyp segmentation for colonoscopy images is of vital importance in clinical practice. It can provide valuable information for colorectal cancer diagnosis and surgery. While existing methods have achieved relatively good performance, polyp segmentation still faces the following challenges: (1) Varying lighting conditions in colonoscopy and differences in polyp locations, sizes, and morphologies. (2) The indistinct boundary between polyps and surrounding tissue. To address these challenges, we propose a Multi-scale information sharing and selection network (MISNet) for polyp segmentation task. We design a Selectively Shared Fusion Module (SSFM) to enforce information sharing and active selection between low-level and high-level features, thereby enhancing model's ability to capture comprehensive information. We then design a Parallel Attention Module (PAM) to enhance model's attention to boundaries, and a Balancing Weight Module (BWM) to facilitate the continuous refinement of boundary segmentation in the bottom-up process. Experiments on five polyp segmentation datasets demonstrate that MISNet successfully improved the accuracy and clarity of segmentation result, outperforming state-of-the-art methods. | computer vision |
Revisiting the Robust Generalization of Adversarial Prompt Tuning | arXiv:2405.11154v1 Announce Type: new
Abstract: Understanding the vulnerability of large-scale pre-trained vision-language models like CLIP against adversarial attacks is key to ensuring zero-shot generalization capacity on various downstream tasks. State-of-the-art defense mechanisms generally adopt prompt learning strategies for adversarial fine-tuning to improve the adversarial robustness of the pre-trained model while keeping the efficiency of adapting to downstream tasks. Such a setup leads to the problem of over-fitting which impedes further improvement of the model's generalization capacity on both clean and adversarial examples. In this work, we propose an adaptive Consistency-guided Adversarial Prompt Tuning (i.e., CAPT) framework that utilizes multi-modal prompt learning to enhance the alignment of image and text features for adversarial examples and leverage the strong generalization of pre-trained CLIP to guide the model-enhancing its robust generalization on adversarial examples while maintaining its accuracy on clean ones. We also design a novel adaptive consistency objective function to balance the consistency of adversarial inputs and clean inputs between the fine-tuning model and the pre-trained model. We conduct extensive experiments across 14 datasets and 4 data sparsity schemes (from 1-shot to full training data settings) to show the superiority of CAPT over other state-of-the-art adaption methods. CAPT demonstrated excellent performance in terms of the in-distribution performance and the generalization under input distribution shift and across datasets. | natural language processing |
Dusk Till Dawn: Self-supervised Nighttime Stereo Depth Estimation using Visual Foundation Models | arXiv:2405.11158v1 Announce Type: new
Abstract: Self-supervised depth estimation algorithms rely heavily on frame-warping relationships, exhibiting substantial performance degradation when applied in challenging circumstances, such as low-visibility and nighttime scenarios with varying illumination conditions. Addressing this challenge, we introduce an algorithm designed to achieve accurate self-supervised stereo depth estimation focusing on nighttime conditions. Specifically, we use pretrained visual foundation models to extract generalised features across challenging scenes and present an efficient method for matching and integrating these features from stereo frames. Moreover, to prevent pixels violating photometric consistency assumption from negatively affecting the depth predictions, we propose a novel masking approach designed to filter out such pixels. Lastly, addressing weaknesses in the evaluation of current depth estimation algorithms, we present novel evaluation metrics. Our experiments, conducted on challenging datasets including Oxford RobotCar and Multi-Spectral Stereo, demonstrate the robust improvements realized by our approach. Code is available at: https://github.com/madhubabuv/dtd | computer vision |
Automated Multi-level Preference for MLLMs | arXiv:2405.11165v1 Announce Type: new
Abstract: Current multimodal Large Language Models (MLLMs) suffer from ``hallucination'', occasionally generating responses that are not grounded in the input images. To tackle this challenge, one promising path is to utilize reinforcement learning from human feedback (RLHF), which steers MLLMs towards learning superior responses while avoiding inferior ones. We rethink the common practice of using binary preferences (\emph{i.e.}, superior, inferior), and find that adopting multi-level preferences (\emph{e.g.}, superior, medium, inferior) is better for two benefits: 1) It narrows the gap between adjacent levels, thereby encouraging MLLMs to discern subtle differences. 2) It further integrates cross-level comparisons (beyond adjacent-level comparisons), thus providing a broader range of comparisons with hallucination examples. To verify our viewpoint, we present the Automated Multi-level Preference (\textbf{AMP}) framework for MLLMs. To facilitate this framework, we first develop an automated dataset generation pipeline that provides high-quality multi-level preference datasets without any human annotators. Furthermore, we design the Multi-level Direct Preference Optimization (MDPO) algorithm to robustly conduct complex multi-level preference learning. Additionally, we propose a new hallucination benchmark, MRHal-Bench. Extensive experiments across public hallucination and general benchmarks, as well as our MRHal-Bench, demonstrate the effectiveness of our proposed method. | natural language processing |
GestFormer: Multiscale Wavelet Pooling Transformer Network for Dynamic Hand Gesture Recognition | arXiv:2405.11180v1 Announce Type: new
Abstract: Transformer model have achieved state-of-the-art results in many applications like NLP, classification, etc. But their exploration in gesture recognition task is still limited. So, we propose a novel GestFormer architecture for dynamic hand gesture recognition. The motivation behind this design is to propose a resource efficient transformer model, since transformers are computationally expensive and very complex. So, we propose to use a pooling based token mixer named PoolFormer, since it uses only pooling layer which is a non-parametric layer instead of quadratic attention. The proposed model also leverages the space-invariant features of the wavelet transform and also the multiscale features are selected using multi-scale pooling. Further, a gated mechanism helps to focus on fine details of the gesture with the contextual information. This enhances the performance of the proposed model compared to the traditional transformer with fewer parameters, when evaluated on dynamic hand gesture datasets, NVidia Dynamic Hand Gesture and Briareo datasets. To prove the efficacy of the proposed model, we have experimented on single as well multimodal inputs such as infrared, normals, depth, optical flow and color images. We have also compared the proposed GestFormer in terms of resource efficiency and number of operations. The source code is available at https://github.com/mallikagarg/GestFormer. | computer vision |
ReasonPix2Pix: Instruction Reasoning Dataset for Advanced Image Editing | arXiv:2405.11190v1 Announce Type: new
Abstract: Instruction-based image editing focuses on equipping a generative model with the capacity to adhere to human-written instructions for editing images. Current approaches typically comprehend explicit and specific instructions. However, they often exhibit a deficiency in executing active reasoning capacities required to comprehend instructions that are implicit or insufficiently defined. To enhance active reasoning capabilities and impart intelligence to the editing model, we introduce ReasonPix2Pix, a comprehensive reasoning-attentive instruction editing dataset. The dataset is characterized by 1) reasoning instruction, 2) more realistic images from fine-grained categories, and 3) increased variances between input and edited images. When fine-tuned with our dataset under supervised conditions, the model demonstrates superior performance in instructional editing tasks, independent of whether the tasks require reasoning or not. The code, model, and dataset will be publicly available. | natural language processing |
Fuse & Calibrate: A bi-directional Vision-Language Guided Framework for Referring Image Segmentation | arXiv:2405.11205v1 Announce Type: new
Abstract: Referring Image Segmentation (RIS) aims to segment an object described in natural language from an image, with the main challenge being a text-to-pixel correlation. Previous methods typically rely on single-modality features, such as vision or language features, to guide the multi-modal fusion process. However, this approach limits the interaction between vision and language, leading to a lack of fine-grained correlation between the language description and pixel-level details during the decoding process. In this paper, we introduce FCNet, a framework that employs a bi-directional guided fusion approach where both vision and language play guiding roles. Specifically, we use a vision-guided approach to conduct initial multi-modal fusion, obtaining multi-modal features that focus on key vision information. We then propose a language-guided calibration module to further calibrate these multi-modal features, ensuring they understand the context of the input sentence. This bi-directional vision-language guided approach produces higher-quality multi-modal features sent to the decoder, facilitating adaptive propagation of fine-grained semantic information from textual features to visual features. Experiments on RefCOCO, RefCOCO+, and G-Ref datasets with various backbones consistently show our approach outperforming state-of-the-art methods. | computer vision |
TriLoRA: Integrating SVD for Advanced Style Personalization in Text-to-Image Generation | arXiv:2405.11236v1 Announce Type: new
Abstract: As deep learning technology continues to advance, image generation models, especially models like Stable Diffusion, are finding increasingly widespread application in visual arts creation. However, these models often face challenges such as overfitting, lack of stability in generated results, and difficulties in accurately capturing the features desired by creators during the fine-tuning process. In response to these challenges, we propose an innovative method that integrates Singular Value Decomposition (SVD) into the Low-Rank Adaptation (LoRA) parameter update strategy, aimed at enhancing the fine-tuning efficiency and output quality of image generation models. By incorporating SVD within the LoRA framework, our method not only effectively reduces the risk of overfitting but also enhances the stability of model outputs, and captures subtle, creator-desired feature adjustments more accurately. We evaluated our method on multiple datasets, and the results show that, compared to traditional fine-tuning methods, our approach significantly improves the model's generalization ability and creative flexibility while maintaining the quality of generation. Moreover, this method maintains LoRA's excellent performance under resource-constrained conditions, allowing for significant improvements in image generation quality without sacrificing the original efficiency and resource advantages. | natural language processing |
Testing the Performance of Face Recognition for People with Down Syndrome | arXiv:2405.11240v1 Announce Type: new
Abstract: The fairness of biometric systems, in particular facial recognition, is often analysed for larger demographic groups, e.g. female vs. male or black vs. white. In contrast to this, minority groups are commonly ignored. This paper investigates the performance of facial recognition algorithms on individuals with Down syndrome, a common chromosomal abnormality that affects approximately one in 1,000 births per year. To do so, a database of 98 individuals with Down syndrome, each represented by at least five facial images, is semi-automatically collected from YouTube. Subsequently, two facial image quality assessment algorithms and five recognition algorithms are evaluated on the newly collected database and on the public facial image databases CelebA and FRGCv2. The results show that the quality scores of facial images for individuals with Down syndrome are comparable to those of individuals without Down syndrome captured under similar conditions. Furthermore, it is observed that face recognition performance decreases significantly for individuals with Down syndrome, which is largely attributed to the increased likelihood of false matches. | computer vision |
Dreamer XL: Towards High-Resolution Text-to-3D Generation via Trajectory Score Matching | arXiv:2405.11252v1 Announce Type: new
Abstract: In this work, we propose a novel Trajectory Score Matching (TSM) method that aims to solve the pseudo ground truth inconsistency problem caused by the accumulated error in Interval Score Matching (ISM) when using the Denoising Diffusion Implicit Models (DDIM) inversion process. Unlike ISM which adopts the inversion process of DDIM to calculate on a single path, our TSM method leverages the inversion process of DDIM to generate two paths from the same starting point for calculation. Since both paths start from the same starting point, TSM can reduce the accumulated error compared to ISM, thus alleviating the problem of pseudo ground truth inconsistency. TSM enhances the stability and consistency of the model's generated paths during the distillation process. We demonstrate this experimentally and further show that ISM is a special case of TSM. Furthermore, to optimize the current multi-stage optimization process from high-resolution text to 3D generation, we adopt Stable Diffusion XL for guidance. In response to the issues of abnormal replication and splitting caused by unstable gradients during the 3D Gaussian splatting process when using Stable Diffusion XL, we propose a pixel-by-pixel gradient clipping method. Extensive experiments show that our model significantly surpasses the state-of-the-art models in terms of visual quality and performance. Code: \url{https://github.com/xingy038/Dreamer-XL}. | computer vision |
HR Human: Modeling Human Avatars with Triangular Mesh and High-Resolution Textures from Videos | arXiv:2405.11270v1 Announce Type: new
Abstract: Recently, implicit neural representation has been widely used to generate animatable human avatars. However, the materials and geometry of those representations are coupled in the neural network and hard to edit, which hinders their application in traditional graphics engines. We present a framework for acquiring human avatars that are attached with high-resolution physically-based material textures and triangular mesh from monocular video. Our method introduces a novel information fusion strategy to combine the information from the monocular video and synthesize virtual multi-view images to tackle the sparsity of the input view. We reconstruct humans as deformable neural implicit surfaces and extract triangle mesh in a well-behaved pose as the initial mesh of the next stage. In addition, we introduce an approach to correct the bias for the boundary and size of the coarse mesh extracted. Finally, we adapt prior knowledge of the latent diffusion model at super-resolution in multi-view to distill the decomposed texture. Experiments show that our approach outperforms previous representations in terms of high fidelity, and this explicit result supports deployment on common renderers. | natural language processing |
Visible and Clear: Finding Tiny Objects in Difference Map | arXiv:2405.11276v1 Announce Type: new
Abstract: Tiny object detection is one of the key challenges in the field of object detection. The performance of most generic detectors dramatically decreases in tiny object detection tasks. The main challenge lies in extracting effective features of tiny objects. Existing methods usually perform generation-based feature enhancement, which is seriously affected by spurious textures and artifacts, making it difficult to make the tiny-object-specific features visible and clear for detection. To address this issue, we propose a self-reconstructed tiny object detection (SR-TOD) framework. We for the first time introduce a self-reconstruction mechanism in the detection model, and discover the strong correlation between it and the tiny objects. Specifically, we impose a reconstruction head in-between the neck of a detector, constructing a difference map of the reconstructed image and the input, which shows high sensitivity to tiny objects. This inspires us to enhance the weak representations of tiny objects under the guidance of the difference maps. Thus, improving the visibility of tiny objects for the detectors. Building on this, we further develop a Difference Map Guided Feature Enhancement (DGFE) module to make the tiny feature representation more clear. In addition, we further propose a new multi-instance anti-UAV dataset, which is called DroneSwarms dataset and contains a large number of tiny drones with the smallest average size to date. Extensive experiments on the DroneSwarms dataset and other datasets demonstrate the effectiveness of the proposed method. The code and dataset will be publicly available. | computer vision |
Motion Avatar: Generate Human and Animal Avatars with Arbitrary Motion | arXiv:2405.11286v1 Announce Type: new
Abstract: In recent years, there has been significant interest in creating 3D avatars and motions, driven by their diverse applications in areas like film-making, video games, AR/VR, and human-robot interaction. However, current efforts primarily concentrate on either generating the 3D avatar mesh alone or producing motion sequences, with integrating these two aspects proving to be a persistent challenge. Additionally, while avatar and motion generation predominantly target humans, extending these techniques to animals remains a significant challenge due to inadequate training data and methods. To bridge these gaps, our paper presents three key contributions. Firstly, we proposed a novel agent-based approach named Motion Avatar, which allows for the automatic generation of high-quality customizable human and animal avatars with motions through text queries. The method significantly advanced the progress in dynamic 3D character generation. Secondly, we introduced a LLM planner that coordinates both motion and avatar generation, which transforms a discriminative planning into a customizable Q&A fashion. Lastly, we presented an animal motion dataset named Zoo-300K, comprising approximately 300,000 text-motion pairs across 65 animal categories and its building pipeline ZooGen, which serves as a valuable resource for the community. See project website https://steve-zeyu-zhang.github.io/MotionAvatar/ | computer vision |
InfRS: Incremental Few-Shot Object Detection in Remote Sensing Images | arXiv:2405.11293v1 Announce Type: new
Abstract: Recently, the field of few-shot detection within remote sensing imagery has witnessed significant advancements. Despite these progresses, the capacity for continuous conceptual learning still poses a significant challenge to existing methodologies. In this paper, we explore the intricate task of incremental few-shot object detection in remote sensing images. We introduce a pioneering fine-tuningbased technique, termed InfRS, designed to facilitate the incremental learning of novel classes using a restricted set of examples, while concurrently preserving the performance on established base classes without the need to revisit previous datasets. Specifically, we pretrain the model using abundant data from base classes and then generate a set of class-wise prototypes that represent the intrinsic characteristics of the data. In the incremental learning stage, we introduce a Hybrid Prototypical Contrastive (HPC) encoding module for learning discriminative representations. Furthermore, we develop a prototypical calibration strategy based on the Wasserstein distance to mitigate the catastrophic forgetting problem. Comprehensive evaluations on the NWPU VHR-10 and DIOR datasets demonstrate that our model can effectively solve the iFSOD problem in remote sensing images. Code will be released. | computer vision |
MediCLIP: Adapting CLIP for Few-shot Medical Image Anomaly Detection | arXiv:2405.11315v1 Announce Type: new
Abstract: In the field of medical decision-making, precise anomaly detection in medical imaging plays a pivotal role in aiding clinicians. However, previous work is reliant on large-scale datasets for training anomaly detection models, which increases the development cost. This paper first focuses on the task of medical image anomaly detection in the few-shot setting, which is critically significant for the medical field where data collection and annotation are both very expensive. We propose an innovative approach, MediCLIP, which adapts the CLIP model to few-shot medical image anomaly detection through self-supervised fine-tuning. Although CLIP, as a vision-language model, demonstrates outstanding zero-/fewshot performance on various downstream tasks, it still falls short in the anomaly detection of medical images. To address this, we design a series of medical image anomaly synthesis tasks to simulate common disease patterns in medical imaging, transferring the powerful generalization capabilities of CLIP to the task of medical image anomaly detection. When only few-shot normal medical images are provided, MediCLIP achieves state-of-the-art performance in anomaly detection and location compared to other methods. Extensive experiments on three distinct medical anomaly detection tasks have demonstrated the superiority of our approach. The code is available at https://github.com/cnulab/MediCLIP. | computer vision |
UPAM: Unified Prompt Attack in Text-to-Image Generation Models Against Both Textual Filters and Visual Checkers | arXiv:2405.11336v1 Announce Type: new
Abstract: Text-to-Image (T2I) models have raised security concerns due to their potential to generate inappropriate or harmful images. In this paper, we propose UPAM, a novel framework that investigates the robustness of T2I models from the attack perspective. Unlike most existing attack methods that focus on deceiving textual defenses, UPAM aims to deceive both textual and visual defenses in T2I models. UPAM enables gradient-based optimization, offering greater effectiveness and efficiency than previous methods. Given that T2I models might not return results due to defense mechanisms, we introduce a Sphere-Probing Learning (SPL) scheme to support gradient optimization even when no results are returned. Additionally, we devise a Semantic-Enhancing Learning (SEL) scheme to finetune UPAM for generating target-aligned images. Our framework also ensures attack stealthiness. Extensive experiments demonstrate UPAM's effectiveness and efficiency. | computer vision |
A Unified Approach Towards Active Learning and Out-of-Distribution Detection | arXiv:2405.11337v1 Announce Type: new
Abstract: When applying deep learning models in open-world scenarios, active learning (AL) strategies are crucial for identifying label candidates from a nearly infinite amount of unlabeled data. In this context, robust out-of-distribution (OOD) detection mechanisms are essential for handling data outside the target distribution of the application. However, current works investigate both problems separately. In this work, we introduce SISOM as the first unified solution for both AL and OOD detection. By leveraging feature space distance metrics SISOM combines the strengths of the currently independent tasks to solve both effectively. We conduct extensive experiments showing the problems arising when migrating between both tasks. In these evaluations SISOM underlined its effectiveness by achieving first place in two of the widely used OpenOOD benchmarks and second place in the remaining one. In AL, SISOM outperforms others and delivers top-1 performance in three benchmarks | computer vision |
EyeFound: A Multimodal Generalist Foundation Model for Ophthalmic Imaging | arXiv:2405.11338v1 Announce Type: new
Abstract: Artificial intelligence (AI) is vital in ophthalmology, tackling tasks like diagnosis, classification, and visual question answering (VQA). However, existing AI models in this domain often require extensive annotation and are task-specific, limiting their clinical utility. While recent developments have brought about foundation models for ophthalmology, they are limited by the need to train separate weights for each imaging modality, preventing a comprehensive representation of multi-modal features. This highlights the need for versatile foundation models capable of handling various tasks and modalities in ophthalmology. To address this gap, we present EyeFound, a multimodal foundation model for ophthalmic images. Unlike existing models, EyeFound learns generalizable representations from unlabeled multimodal retinal images, enabling efficient model adaptation across multiple applications. Trained on 2.78 million images from 227 hospitals across 11 ophthalmic modalities, EyeFound facilitates generalist representations and diverse multimodal downstream tasks, even for detecting challenging rare diseases. It outperforms previous work RETFound in diagnosing eye diseases, predicting systemic disease incidents, and zero-shot multimodal VQA. EyeFound provides a generalizable solution to improve model performance and lessen the annotation burden on experts, facilitating widespread clinical AI applications from retinal imaging. | natural language processing |
City-Scale Multi-Camera Vehicle Tracking System with Improved Self-Supervised Camera Link Model | arXiv:2405.11345v1 Announce Type: new
Abstract: Multi-Target Multi-Camera Tracking (MTMCT) has broad applications and forms the basis for numerous future city-wide systems (e.g. traffic management, crash detection, etc.). However, the challenge of matching vehicle trajectories across different cameras based solely on feature extraction poses significant difficulties. This article introduces an innovative multi-camera vehicle tracking system that utilizes a self-supervised camera link model. In contrast to related works that rely on manual spatial-temporal annotations, our model automatically extracts crucial multi-camera relationships for vehicle matching. The camera link is established through a pre-matching process that evaluates feature similarities, pair numbers, and time variance for high-quality tracks. This process calculates the probability of spatial linkage for all camera combinations, selecting the highest scoring pairs to create camera links. Our approach significantly improves deployment times by eliminating the need for human annotation, offering substantial improvements in efficiency and cost-effectiveness when it comes to real-world application. This pairing process supports cross camera matching by setting spatial-temporal constraints, reducing the searching space for potential vehicle matches. According to our experimental results, the proposed method achieves a new state-of-the-art among automatic camera-link based methods in CityFlow V2 benchmarks with 61.07% IDF1 Score. | computer vision |
PlantTracing: Tracing Arabidopsis Thaliana Apex with CenterTrack | arXiv:2405.11351v1 Announce Type: new
Abstract: This work applies an encoder-decoder-based machine learning network to detect and track the motion and growth of the flowering stem apex of Arabidopsis Thaliana. Based on the CenterTrack, a machine learning back-end network, we trained a model based on ten time-lapsed labeled videos and tested against three videos. | computer vision |
The First Swahili Language Scene Text Detection and Recognition Dataset | arXiv:2405.11437v1 Announce Type: new
Abstract: Scene text recognition is essential in many applications, including automated translation, information retrieval, driving assistance, and enhancing accessibility for individuals with visual impairments. Much research has been done to improve the accuracy and performance of scene text detection and recognition models. However, most of this research has been conducted in the most common languages, English and Chinese. There is a significant gap in low-resource languages, especially the Swahili Language. Swahili is widely spoken in East African countries but is still an under-explored language in scene text recognition. No studies have been focused explicitly on Swahili natural scene text detection and recognition, and no dataset for Swahili language scene text detection and recognition is publicly available. We propose a comprehensive dataset of Swahili scene text images and evaluate the dataset on different scene text detection and recognition models. The dataset contains 976 images collected in different places and under various circumstances. Each image has its annotation at the word level. The proposed dataset can also serve as a benchmark dataset specific to the Swahili language for evaluating and comparing different approaches and fostering future research endeavors. The dataset is available on GitHub via this link: https://github.com/FadilaW/Swahili-STR-Dataset | natural language processing |
Unifying 3D Vision-Language Understanding via Promptable Queries | arXiv:2405.11442v1 Announce Type: new
Abstract: A unified model for 3D vision-language (3D-VL) understanding is expected to take various scene representations and perform a wide range of tasks in a 3D scene. However, a considerable gap exists between existing methods and such a unified model, due to the independent application of representation and insufficient exploration of 3D multi-task training. In this paper, we introduce PQ3D, a unified model capable of using Promptable Queries to tackle a wide range of 3D-VL tasks, from low-level instance segmentation to high-level reasoning and planning. This is achieved through three key innovations: (1) unifying various 3D scene representations (i.e., voxels, point clouds, multi-view images) into a shared 3D coordinate space by segment-level grouping, (2) an attention-based query decoder for task-specific information retrieval guided by prompts, and (3) universal output heads for different tasks to support multi-task training. Tested across ten diverse 3D-VL datasets, PQ3D demonstrates impressive performance on these tasks, setting new records on most benchmarks. Particularly, PQ3D improves the state-of-the-art on ScanNet200 by 1.8% (AP), ScanRefer by 5.4% (acc@0.5), Multi3DRefer by 11.7% (F1@0.5), and Scan2Cap by 13.4% (CIDEr@0.5). Moreover, PQ3D supports flexible inference with individual or combined forms of available 3D representations, e.g., solely voxel input. | computer vision |
Cross-Domain Knowledge Distillation for Low-Resolution Human Pose Estimation | arXiv:2405.11448v1 Announce Type: new
Abstract: In practical applications of human pose estimation, low-resolution inputs frequently occur, and existing state-of-the-art models perform poorly with low-resolution images. This work focuses on boosting the performance of low-resolution models by distilling knowledge from a high-resolution model. However, we face the challenge of feature size mismatch and class number mismatch when applying knowledge distillation to networks with different input resolutions. To address this issue, we propose a novel cross-domain knowledge distillation (CDKD) framework. In this framework, we construct a scale-adaptive projector ensemble (SAPE) module to spatially align feature maps between models of varying input resolutions. It adopts a projector ensemble to map low-resolution features into multiple common spaces and adaptively merges them based on multi-scale information to match high-resolution features. Additionally, we construct a cross-class alignment (CCA) module to solve the problem of the mismatch of class numbers. By combining an easy-to-hard training (ETHT) strategy, the CCA module further enhances the distillation performance. The effectiveness and efficiency of our approach are demonstrated by extensive experiments on two common benchmark datasets: MPII and COCO. The code is made available in supplementary material. | natural language processing |
AdaAugment: A Tuning-Free and Adaptive Approach to Enhance Data Augmentation | arXiv:2405.11467v1 Announce Type: new
Abstract: Data augmentation (DA) is widely employed to improve the generalization performance of deep models. However, most existing DA methods use augmentation operations with random magnitudes throughout training. While this fosters diversity, it can also inevitably introduce uncontrolled variability in augmented data, which may cause misalignment with the evolving training status of the target models. Both theoretical and empirical findings suggest that this misalignment increases the risks of underfitting and overfitting. To address these limitations, we propose AdaAugment, an innovative and tuning-free Adaptive Augmentation method that utilizes reinforcement learning to dynamically adjust augmentation magnitudes for individual training samples based on real-time feedback from the target network. Specifically, AdaAugment features a dual-model architecture consisting of a policy network and a target network, which are jointly optimized to effectively adapt augmentation magnitudes. The policy network optimizes the variability within the augmented data, while the target network utilizes the adaptively augmented samples for training. Extensive experiments across benchmark datasets and deep architectures demonstrate that AdaAugment consistently outperforms other state-of-the-art DA methods in effectiveness while maintaining remarkable efficiency. | computer vision |