Memes have evolved as a prevalent medium for diverse communication, ranging from humour to propaganda. With the rising popularity of image-focused content, there is a growing need to explore its potential harm from different aspects. Previous studies have analyzed memes in closed settings - detecting harm, applying semantic labels, and offering natural language explanations. To extend this research, we introduce MemeMQA, a multimodal question-answering framework aiming to solicit accurate responses to structured questions while providing coherent explanations. We curate MemeMQACorpus, a new dataset featuring 1,880 questions related to 1,122 memes with corresponding answer-explanation pairs. We further propose ARSENAL, a novel two-stage multimodal framework that leverages the reasoning capabilities of LLMs to address MemeMQA. We benchmark MemeMQA using competitive baselines and demonstrate its superiority - ~18% enhanced answer prediction accuracy and distinct text generation lead across various metrics measuring lexical and semantic alignment over the best baseline. We analyze ARSENAL’s robustness through diversification of question-set, confounder-based evaluation regarding MemeMQA’s generalizability, and modality-specific assessment, enhancing our understanding of meme interpretation in the multimodal communication landscape.
论文检索
输入标题、作者或关键词，从 9,460 篇学术成果中精准定位
Reformulating Domain Adaptation of Large Language Models as Adapt-Retrieve-Revise: A Case Study on Chinese Legal Domain
PDF ↗While large language models (LLMs) like GPT-4 have recently demonstrated astonishing zero-shot capabilities in general domain tasks, they often generate content with hallucinations in specific domains such as Chinese law, hindering their application in these areas. This is typically due to the absence of training data that encompasses such a specific domain, preventing GPT-4 from acquiring in-domain knowledge. A pressing challenge is that it’s not plausible to continue training LLMs of the GPT-4’s scale on in-domain data.This paper introduces a simple yet effective domain adaptation framework for GPT-4 by reformulating generation as an adapt-retrieve-revise process. The initial step is to adapt an affordable 7B LLM to the Chinese legal domain by continuing learning in-domain data. When solving an in-domain task, we leverage the adapted LLM to generate a draft answer given a task query. Then, the draft answer will be used to retrieve supporting evidence candidates from an external in-domain knowledge base. Finally, the draft answer and retrieved evidence are concatenated into a whole prompt to let GPT-4 assess the evidence and revise the draft answer to generate the final answer. Our proposal combines the advantages of the efficiency of adapting a smaller 7B model with the evidence-assessing capability of GPT-4 and effectively prevents GPT-4 from generating hallucinatory content. In the zero-shot setting of four Chinese legal tasks, our method improves the average score by +33.6 points, compared to GPT-4 direct generation. When compared to two stronger retrieval-based baselines, our method outperforms them by +17.0 and +23.5.
Multimodal knowledge editing represents a critical advancement in enhancing the capabilities of Multimodal Large Language Models (MLLMs). Despite its potential, current benchmarks predominantly focus on coarse-grained knowledge, leaving the intricacies of fine-grained (FG) multimodal entity knowledge largely unexplored. This gap presents a notable challenge, as FG entity recognition is pivotal for the practical deployment and effectiveness of MLLMs in diverse real-world scenarios. To bridge this gap, we introduce MIKE, a comprehensive benchmark and dataset specifically designed for the FG multimodal entity knowledge editing. MIKE encompasses a suite of tasks tailored to assess different perspectives, including Vanilla Name Answering, Entity-Level Caption, and Complex-Scenario Recognition. In addition, a new form of knowledge editing, Multi-step Editing, is introduced to evaluate the editing efficiency. Through our extensive evaluations, we demonstrate that the current state-of-the-art methods face significant challenges in tackling our proposed benchmark, underscoring the complexity of FG knowledge editing in MLLMs. Our findings spotlight the urgent need for novel approaches in this domain, setting a clear agenda for future research and development efforts within the community.
Reinforcement Learning from Human Feedback (RLHF) has been a crucial component in the recent success of Large Language Models. However, RLHF is know to exploit biases in human preferences, such as verbosity. A well-formatted and eloquent answer is often more highly rated by users, even when it is less helpful and objective. A number of approaches have been developed to control those biases in the classical RLHF literature, but the problem remains relatively under-explored for Direct Alignment Algorithms such as Direct Preference Optimization (DPO). Unlike classical RLHF, DPO does not train a separate reward model or use reinforcement learning directly, so previous approaches developed to control verbosity cannot be directly applied to this setting. Our work makes several contributions. For the first time, we study the length problem in the DPO setting, showing significant exploitation in DPO and linking it to out-of-distribution bootstrapping. We then develop a principled but simple regularization strategy that prevents length exploitation, while still maintaining improvements in model quality. We demonstrate these affects across datasets on summarization and dialogue, where we achieve up to 20% improvement in win rates when controlling for length, despite the GPT4 judge’s well-known verbosity bias.
MLeVLM: Improve Multi-level Progressive Capabilities based on Multimodal Large Language Model for Medical Visual Question Answering
PDF ↗Medical visual question answering (MVQA) requires in-depth understanding of medical images and questions to provide reliable answers. We summarize multi-level progressive capabilities that models need to focus on in MVQA: recognition, details, diagnosis, knowledge, and reasoning. Existing MVQA models tend to ignore the above capabilities due to unspecific data and plain architecture. To address these issues, this paper proposes Multi-level Visual Language Model (MLeVLM) for MVQA. On the data side, we construct a high-quality multi-level instruction dataset MLe-VQA via GPT-4, which covers multi-level questions and answers as well as reasoning processes from visual clues to semantic cognition. On the architecture side, we propose a multi-level feature alignment module, including attention-based token selector and context merger, which can efficiently align features at different levels from visual to semantic. To better evaluate the model’s capabilities, we manually construct a multi-level MVQA evaluation benchmark named MLe-Bench. Extensive experiments demonstrate the effectiveness of our constructed multi-level instruction dataset and the multi-level feature alignment module. It also proves that MLeVLM outperforms existing medical multimodal large language models.
Large language models (LLMs) predominantly employ decoder-only transformer architectures, necessitating the retention of keys/values information for historical tokens to provide contextual information and avoid redundant computation. However, the substantial size and parameter volume of these LLMs require massive GPU memory. This memory demand increases with the length of the input text, leading to an urgent need for more efficient methods of information storage and processing. This study introduces Anchor-based LLMs (AnLLMs), which utilize an innovative anchor-based self-attention network (AnSAN) and also an anchor-based inference strategy. This approach enables LLMs to compress sequence information into an anchor token, reducing the keys/values cache and enhancing inference efficiency. Experiments on question-answering benchmarks reveal that AnLLMs maintain similar accuracy levels while achieving up to 99% keys/values cache reduction and up to 3.5 times faster inference. Despite a minor compromise in accuracy, the substantial enhancements of AnLLMs employing the AnSAN technique in resource utilization and computational efficiency underscore their potential for practical LLM applications.
Your Co-Workers Matter: Evaluating Collaborative Capabilities of Language Models in Blocks World
PDF ↗Language agents that interact with the world on their own have great potential for automating digital tasks. While large language model (LLM) agents have made progress in understanding and executing tasks such as textual games and webpage control, many real-world tasks also require collaboration with humans or other LLMs in equal roles, which involves intent understanding, task coordination, and communication. To test LLM’s ability to collaborate, we design a blocks-world environment, where two agents, each having unique goals and skills, build a target structure together. To complete the goals, they can act in the world and communicate in natural language. Under this environment, we design increasingly challenging settings to evaluate different collaboration perspectives, from independent to more complex, dependent tasks. We further adopt chain-of-thought prompts that include intermediate reasoning steps to model the partner’s state and identify and correct execution errors. Both human-machine and machine-machine experiments show that LLM agents have strong grounding capacities, and our approach significantly improves the evaluation metric.
Contrastive cross-modal models such as CLIP and CLAP aid various vision-language (VL) and audio-language (AL) tasks. However, there has been limited investigation of and improvement in their language encoder – the central component of encoding natural language descriptions of image/audio into vector representations. We extensively evaluate how unsupervised and supervised sentence embedding training affect language encoder quality and cross-modal task performance. In VL pretraining, we found that sentence embedding training enhances language encoder quality and aids in cross-modal tasks, improving contrastive VL models such as CyCLIP. Sentence embedding training benefits AL tasks when the amount of training data is large. We analyze the representation spaces to understand the strengths of sentence embedding training, and find that it improves text-space uniformity, at the cost of decreased cross-modal alignment.
Deep neural networks exhibit vulnerability to word-level adversarial attacks in natural language processing. Most of these attack methods adopt synonymous substitutions to perturb original samples for crafting adversarial examples while attempting to maintain semantic consistency with the originals. Some of them claim that they could achieve over 90% attack success rate, thereby raising serious safety concerns. However, our investigation reveals that many purportedly successful adversarial examples are actually invalid due to significant changes in semantic meanings compared to their originals. Even when equipped with semantic constraints such as BERTScore, existing attack methods can generate up to 87.9% invalid adversarial examples. Building on this insight, we first curate a 13K dataset for adversarial validity evaluation with the help of GPT-4. Then, an open-source large language model is fine-tuned to offer an interpretable validity score for assessing the semantic consistency between original and adversarial examples. Finally, this validity score can serve as a guide for existing adversarial attack methods to generate valid adversarial examples. Comprehensive experiments demonstrate the effectiveness of our method in evaluating and refining the quality of adversarial examples.
In this paper, we begin by illustrating that, when presented with a query, Large Language Models (LLMs) capable of providing accurate responses tend to exhibit a more uniform probability distribution compared to their less proficient counterparts. Building upon this observation, we introduce a novel self-assessment criterion termed ProbDiff for evaluating the performance of diverse LLMs. This method eliminates the need for training an additional evaluation model or relying on external proprietary models such as GPT-4 as a judger. Instead, it solely relies on the LLMs under evaluation to compute the probability discrepancy between the original response generation and its revised versions. A higher discrepancy in two LLMs for the same query suggests a relatively weaker ability. We discover that ProbDiff yields comparable results to mainstream GPT-4-based evaluations on various scenarios including NLG tasks like translation and summarization, as well as LLM evaluation benchmarks such as AlignBench, MT-Bench, and AlpacaEval, across LLMs of different sizes.
Towards Tracing Trustworthiness Dynamics: Revisiting Pre-training Period of Large Language Models
PDF ↗Ensuring the trustworthiness of large language models (LLMs) is crucial. Most studies concentrate on fully pre-trained LLMs to better understand and improve LLMs’ trustworthiness. In this paper, to reveal the untapped potential of pre-training, we pioneer the exploration of LLMs’ trustworthiness during this period, focusing on five key dimensions: reliability, privacy, toxicity, fairness, and robustness. To begin with, we apply linear probing to LLMs. The high probing accuracy suggests that LLMs in early pre-training can already distinguish concepts in each trustworthiness dimension. Therefore, to further uncover the hidden possibilities of pre-training, we extract steering vectors from a LLM’s pre-training checkpoints to enhance the LLM’s trustworthiness. Finally, inspired by the theoretical result that mutual information estimation is bounded by linear probing accuracy, we also probe LLMs with mutual information to investigate the dynamics of trustworthiness during pre-training. We are the first to observe a similar two-phase phenomenon: fitting and compression. This research provides an initial exploration of trustworthiness modeling during LLM pre-training, seeking to unveil new insights and spur further developments in the field.
The success of state-of-the-art Natural Language Processing (NLP) systems heavily depends on deep neural networks, which excel in various tasks through strong data fitting and latent feature modeling abilities. However, certain challenges linked to deep neural networks and supervised deep learning deserve considerations, e.g., extensive computing resources, knowledge forgetting, etc. Previous research attempted to tackle these challenges individually through irrelative techniques. However, they do not instigate fundamental shifts in the learning paradigm. In this work, we propose a novel neurosymbolic method for sentiment analysis to tackle these issues. We also propose a novel sentiment-pragmatic knowledge base that places emphasis on human subjectivity within varying domain annotations. We conducted extensive experiments to show that our neurosymbolic framework for sentiment analysis stands out for its lightweight nature, robustness across domains and languages, efficient few-shot training, and rapid convergence.
Unraveling and Mitigating Retriever Inconsistencies in Retrieval-Augmented Large Language Models
PDF ↗Although Retrieval-Augmented Large Language Models (RALMs) demonstrate their superiority in terms of factuality, they do not consistently outperform the original retrieval-free Language Models (LMs). Our experiments reveal that this example-level performance inconsistency exists not only between retrieval-augmented and retrieval-free LM but also among different retrievers. To understand this phenomenon, we investigate the degeneration behavior of RALMs and theoretically decompose it into four categories. Further analysis based on our decomposition reveals that the innate difference in knowledge sources and the unpredictable degeneration of the reader model contribute most to the inconsistency. Drawing from our analysis, we introduce Ensemble of Retrievers (EoR), a trainable framework that can adaptively retrieve from different knowledge sources and effectively decrease unpredictable reader errors. Our experiments on Open Domain Question Answering show that EoR substantially improves performance over the RALM with a single retriever by considerably reducing inconsistent behaviors.
Relational concepts are indeed foundational to the structure of knowledge representation, as they facilitate the association between various entity concepts, allowing us to express and comprehend complex world knowledge.By expressing relational concepts in natural language prompts, people can effortlessly interact with large language models (LLMs) and recall desired factual knowledge. However, the process of knowledge recall lacks interpretability, and representations of relational concepts within LLMs remain unknown to us. In this paper, we identify hidden states that can express entity and relational concepts through causal mediation analysis in fact recall processes. Our finding reveals that at the last token position of the input prompt, there are hidden states that solely express the causal effects of relational concepts. Based on this finding, we assume that these hidden states can be treated as relational representations and we can successfully extract them from LLMs. The experimental results demonstrate high credibility of the relational representations: they can be flexibly transplanted into other fact recall processes, and can also be used as robust entity connectors. Moreover, we also show that the relational representations exhibit significant potential for controllable fact recall through relation rewriting.
Incorporating Syntax and Lexical Knowledge to Multilingual Sentiment Classification on Large Language Models
PDF ↗This paper exploits a sentiment extractor supported by syntactic and lexical resources to enhance multilingual sentiment classification solved through the generative approach, without retraining LLMs. By adding external information of words and phrases that have positive/negative polarities, the multilingual sentiment classification error was reduced by up to 33 points, and the combination of two approaches performed best especially in high-performing pairs of LLMs and languages.
Unveiling the Truth and Facilitating Change: Towards Agent-based Large-scale Social Movement Simulation
PDF ↗Social media has emerged as a cornerstone of social movements, wielding significant influence in driving societal change. Simulating the response of the public and forecasting the potential impact has become increasingly important. However, existing methods for simulating such phenomena encounter challenges concerning their efficacy and efficiency in capturing the behaviors of social movement participants. In this paper, we introduce a hybrid framework for social media user simulation, wherein users are categorized into two types. Core users are driven by Large Language Models, while numerous ordinary users are modeled by deductive agent-based models. We further construct a Twitter-like environment to replicate their response dynamics following trigger events. Subsequently, we develop a multi-faceted benchmark SoMoSiMu-Bench for evaluation and conduct comprehensive experiments across real-world datasets. Experimental results demonstrate the effectiveness and flexibility of our method.
Aspect-based sentiment analysis (ABSA) identifies sentiment information related to specific aspects and provides deeper market insights to businesses and organizations. With the emergence of large language models (LMs), recent studies have proposed using fixed examples for instruction tuning to reformulate ABSA as a generation task. However, the performance is sensitive to the selection of in-context examples; several retrieval methods are based on surface similarity and are independent of the LM generative objective. This study proposes an instruction learning method with retrieval-based example ranking for ABSA tasks. For each target sample, an LM was applied as a scorer to estimate the likelihood of the output given the input and a candidate example as the prompt, and training examples were labeled as positive or negative by ranking the scores. An alternating training schema is proposed to train both the retriever and LM. Instructional prompts can be constructed using high-quality examples. The LM is used for both scoring and inference, improving the generation efficiency without incurring additional computational costs or training difficulties. Extensive experiments on three ABSA subtasks verified the effectiveness of the proposed method, demonstrating its superiority over various strong baseline models. Code and data are released at https://github.com/zgMin/IT-RER-ABSA.
Chain-of-Question: A Progressive Question Decomposition Approach for Complex Knowledge Base Question Answering
PDF ↗Complex KBQA leverages the knowledge base (KB) to answer complex natural questions involving complicated semantics like multi-hop reasoning. Existing methods involve a question decomposition process, i.e., breaking a complex question into several simpler sub-questions, to assist obtaining logical forms for querying the KB. However, existing question decomposition process derives all sub-questions directly according to the original question, resulting in limitations when one sub-question relies on the answer from a previous one. In this work, we propose Chain-of-Question, a progressive question decomposition approach to address complex KBQA challenges. First, inspired by chain-of-thought, we design a prompt to guide LLM to sequentially decompose multiple semantically clear sub-questions and provide corresponding reference answers, where each step of the decomposition relies on the previous results. Next, we utilize the decomposition result to select relevant patterns (relation-entity pairs) as accurate and faithful auxiliary information for the following logical form generation. Finally, we jointly perform logical form generation and answer prediction, utilizing the predicted answer to supplement non-executable logical forms. Experimental results demonstrate that our method achieves state-of-the-art performance on multiple datasets.
Large Language Models (LLMs) often struggle with hallucinations and outdated information. To address this, Information Retrieval (IR) systems can be employed to augment LLMs with up-to-date knowledge. However, existing IR techniques contain deficiencies, posing a performance bottleneck. Given the extensive array of IR systems, combining diverse approaches presents a viable strategy. Nevertheless, prior attempts have yielded restricted efficacy. In this work, we propose an approach that leverages learning-to-rank techniques to combine heterogeneous IR systems. We demonstrate the method on two Retrieval Question Answering (ReQA) tasks. Our empirical findings exhibit a significant performance enhancement, outperforming previous approaches and achieving state-of-the-art results on ReQA SQuAD.
RA-ISF: Learning to Answer and Understand from Retrieval Augmentation via Iterative Self-Feedback
PDF ↗Large language models (LLMs) demonstrate exceptional performance in numerous tasks but still heavily rely on knowledge stored in their parameters. Moreover, updating this knowledge incurs high training costs. Retrieval-augmented generation (RAG) methods address this issue by integrating external knowledge. The model can answer questions it couldn’t previously by retrieving knowledge relevant to the query. This approach improves performance in certain scenarios for specific tasks. However, if irrelevant texts are retrieved, it may impair model performance. In this paper, we propose Retrieval Augmented Iterative Self-Feedback (RA-ISF), a framework that iteratively decomposes tasks and processes them in three submodules to enhance the model’s problem-solving capabilities. Experiments show that our method outperforms existing benchmarks, performing well on models like GPT3.5, Llama2, significantly enhancing factual reasoning capabilities and reducing hallucinations.