Phrases are fundamental linguistic units through which humans convey semantics. This study critically examines the capacity of API-based large language models (LLMs) to comprehend phrase semantics, utilizing three human-annotated datasets. We assess the performance of LLMs in executing phrase semantic reasoning tasks guided by natural language instructions and explore the impact of common prompting techniques, including few-shot demonstrations and Chain-of-Thought reasoning. Our findings reveal that LLMs greatly outperform traditional embedding methods across the datasets; however, they do not show a significant advantage over fine-tuned methods. The effectiveness of advanced prompting strategies shows variability. We conduct detailed error analyses to interpret the limitations faced by LLMs in comprehending phrase semantics. Code and data can be found at https://github.com/memray/llm_phrase_semantics/.
论文检索
输入标题、作者或关键词，从 2,271 篇学术成果中精准定位
Despite the outstanding performance, Neural Architecture Search (NAS) is criticized for massive computation. Recently, Zero-shot NAS has emerged as a promising approach by exploiting Zero-cost (ZC) proxies, which markedly reduce computational demands. Despite this, existing ZC proxies heavily rely on expert knowledge and incur significant trial-and-error costs. Particularly in NLP tasks, most existing ZC proxies fail to surpass the performance of the naive baseline. To address these challenges, we introduce a novel framework, LPZero, which is the first to automatically design zero-cost (ZC) proxies for various tasks, achieving higher ranking consistency than human-designed proxies. Specifically, we model the ZC proxy as a symbolic equation and incorporate a unified proxy search space that encompasses existing ZC proxies, which are composed of a predefined set of mathematical symbols. To heuristically search for the best ZC proxy, LPZero incorporates genetic programming to find the optimal symbolic composition. We propose a Predictive-Pruning Strategy (PPS), which preemptively eliminates unpromising proxies, thereby mitigating the risk of proxy degradation. Extensive experiments on FlexiBERT, GPT-2, and LLaMA-7B demonstrate LPZero’s superior ranking ability and performance on downstream tasks compared to current approaches.
Infrared-LLaVA: Enhancing Understanding of Infrared Images in Multi-Modal Large Language Models
PDF ↗Expanding the understanding capabilities of multi-modal large language models (MLLMs) for infrared modality is a challenge due to the single-modality nature and limited amount of training data. Existing methods typically construct a uniform embedding space for cross-modal alignment and leverage abundant visual image data to indirectly understand infrared images. However, they ignore the supervisory signals of infrared-modality-specific attributes, which may lead to biased understanding of infrared images. To address this issue, we propose a debating multi-agent generation system which transfers knowledge from visible images to generate infrared image-text pairs and infrared instruction data. Moreover, we construct an infrared question-answering benchmark based on common infrared tasks. Experimental results from incremental fine-tuning on existing models and our Infrared-LLaVA-7B trained from scratch on infrared data demonstrate the effectiveness of the generated data and the feasibility of the generation approach.
Characters are integral to long-form narratives, but are poorly understood by existing story analysis and generation systems. While prior work has simplified characters via graph-based methods and brief character descriptions, we aim to better tackle the problem of representing complex characters by taking inspiration from advice given to professional writers. We propose CHIRON, a new ‘character sheet’ based representation that organizes and filters textual information about characters. We construct CHIRON sheets in two steps: a Generation Module that prompts an LLM for character information via question-answering and a Validation Module that uses automated reasoning and a domain-specific entailment model to eliminate false facts about a character. We validate CHIRON via the downstream task of masked-character prediction, where our experiments show CHIRON is better and more flexible than comparable summary-based baselines. We also show that metrics derived from CHIRON can be used to automatically infer character-centricity in stories, and that these metrics align with human judgments.
AlignSum: Data Pyramid Hierarchical Fine-tuning for Aligning with Human Summarization Preference
PDF ↗Text summarization tasks commonly employ Pre-trained Language Models (PLMs) to fit diverse standard datasets. While these PLMs excel in automatic evaluations, they frequently underperform in human evaluations, indicating a deviation between their generated summaries and human summarization preferences. This discrepancy is likely due to the low quality of fine-tuning datasets and the limited availability of high-quality human-annotated data that reflect true human preference. To address this challenge, we introduce a novel human summarization preference alignment framework AlignSum. This framework consists of three parts: Firstly, we construct a Data Pymarid with extractive, abstractive, and human-annotated summary data. Secondly, we conduct the Gaussian Resampling to remove summaries with extreme lengths. Finally, we implement the two-stage hierarchical fine-tuning with Data Pymarid after Gaussian Resampling. We apply AlignSum to PLMs on the human-annotated CNN/DailyMail and BBC XSum datasets. Experiments show that with AlignSum, PLMs like BART-Large surpass 175B GPT-3 in both automatic and human evaluations. This demonstrates that AlignSum significantly enhances the alignment of language models with human summarization preferences.
Diffusion models exhibit promising capacity for generating high-quality text. However, owing to the curved nature of generation path, they necessitate traversing numerous steps to guarantee the text quality. In this paper, we propose an efficient model FMSeq, which utilizes flow matching to straighten the generation path, thereby enabling fast sampling for diffusion-based seq2seq text generation. Specifically, we construct transport flow only on the target sequences to adapt the diffusion-based model with flow matching. Furthermore, we explore different settings and identify target-parameterization, self-conditioning and time-difference as three effective techniques to improve the generation quality under a few steps. Experiments on four popular tasks demonstrate that FMSeq generates texts of comparable quality to the SOTA diffusion-based DiffuSeq in just 10 steps, achieving a 200-fold speedup.
TRACE the Evidence: Constructing Knowledge-Grounded Reasoning Chains for Retrieval-Augmented Generation
PDF ↗Retrieval-augmented generation (RAG) offers an effective approach for addressing question answering (QA) tasks. However, the imperfections of the retrievers in RAG models often result in the retrieval of irrelevant information, which could introduce noise and degrade the performance, especially when handling multi-hop questions that require multiple steps of reasoning. To enhance the multi-hop reasoning ability of RAG models, we propose TRACE. TRACE constructs knowledge-grounded reasoning chains, which are a series of logically connected knowledge triples, to identify and integrate supporting evidence from the retrieved documents for answering questions. Specifically, TRACE employs a KG Generator to create a knowledge graph (KG) from the retrieved documents, and then uses a novel Autoregressive Reasoning Chain Constructor to build reasoning chains. Experimental results on three multi-hop QA datasets show that TRACE achieves an average performance improvement of up to 14.03% compared to using all the retrieved documents. Moreover, the results indicate that using reasoning chains as context, rather than the entire documents, is often sufficient to correctly answer questions.
Language style is necessary for AI systems to accurately understand and generate diverse human language. However, previous text style transfer primarily focused on sentence-level data-driven approaches, limiting exploration of potential problems in large language models (LLMs) and the ability to meet complex application needs. To overcome these limitations, we introduce a novel task called Public-Speaking Style Transfer (PSST), which aims to simulate humans to transform passage-level, official texts into a public-speaking style. Grounded in the analysis of real-world data from a linguistic perspective, we decompose public-speaking style into key sub-styles to pose challenges and quantify the style modeling capability of LLMs. For such intricate text style transfer, we further propose a fine-grained evaluation framework to analyze the characteristics and identify the problems of stylized texts. Comprehensive experiments suggest that current LLMs struggle to generate public speaking texts that align with human preferences, primarily due to excessive stylization and loss of semantic information. We will release our data, code, and model upon acceptance.
Traditional keyphrase prediction methods predict a single set of keyphrases per document, failing to cater to the diverse needs of users and downstream applications. To bridge the gap, we introduce on-demand keyphrase generation, a novel paradigm that requires keyphrases that conform to specific high-level goals or intents. For this task, we present MetaKP, a large-scale benchmark comprising four datasets, 7500 documents, and 3760 goals across news and biomedical domains with human-annotated keyphrases. Leveraging MetaKP, we design both supervised and unsupervised methods, including a multi-task fine-tuning approach and a self-consistency prompting method with large language models. The results highlight the challenges of supervised fine-tuning, whose performance is not robust to distribution shifts. By contrast, the proposed self-consistency prompting approach greatly improves the performance of large language models, enabling GPT-4o to achieve 0.548 SemF1, surpassing the performance of a fully fine-tuned BART-base model. Finally, we demonstrate the potential of our method to serve as a general NLP infrastructure, exemplified by its application in epidemic event detection from social media.
Large vision-language models (LVLMs) are prone to hallucinations, where certain contextual cues in an image can trigger the language module to produce overconfident and incorrect reasoning about abnormal or hypothetical objects. While some benchmarks have been developed to investigate LVLM hallucinations, they often rely on hand-crafted corner cases whose failure patterns may not generalize well. Additionally, fine-tuning on these examples could undermine their validity. To address this, we aim to scale up the number of cases through an automated approach, reducing human bias in crafting such corner cases. This motivates the development of AutoHallusion, the first automated benchmark generation approach that employs several key strategies to create a diverse range of hallucination examples. Our generated visual-question pairs pose significant challenges to LVLMs, requiring them to overcome contextual biases and distractions to arrive at correct answers. AutoHallusion enables us to create new benchmarks at the minimum cost and thus overcomes the fragility of hand-crafted benchmarks. It also reveals common failure patterns and reasons, providing key insights to detect, avoid, or control hallucinations. Comprehensive evaluations of top-tier LVLMs, e.g., GPT-4V(ision), Gemini Pro Vision, Claude 3, and LLaVA-1.5, show a 97.7% and 98.7% success rate of hallucination induction on synthetic and real-world datasets of AutoHallusion, paving the way for a long battle against hallucinations. The codebase and data can be accessed at https://github.com/wuxiyang1996/AutoHallusion
While LLMs have emerged as performant architectures for reasoning tasks, their compositional generalization capabilities have been questioned. In this work, we introduce a Compositional Generalization Challenge for Graph-based Commonsense Reasoning (CGGC) that goes beyond previous evaluations that are based on sequences or tree structures – and instead involves a reasoning graph: It requires models to generate a natural sentence based on given concepts and a corresponding reasoning graph, where the presented graph involves a previously unseen combination of relation types. To master this challenge, models need to learn how to reason over relation tupels within the graph, and how to compose them when conceptualizing a verbalization. We evaluate seven well-known LLMs using in-context learning and find that performant LLMs still struggle in compositional generalization. We investigate potential causes of this gap by analyzing the structures of reasoning graphs, and find that different structures present varying levels of difficulty for compositional generalization. Arranging the order of demonstrations according to the structures’ difficulty shows that organizing samples in an easy-to-hard schema enhances the compositional generalization ability of LLMs.
Language models are capable of memorizing detailed patterns and information, leading to a double-edged effect: they achieve impressive modeling performance on downstream tasks with the stored knowledge but also raise significant privacy concerns. Traditional differential privacy based training approaches offer robust safeguards by employing a uniform noise distribution across all parameters. However, this overlooks the distinct sensitivities and contributions of individual parameters in privacy protection and often results in suboptimal models. To address these limitations, we propose ANADP, a novel algorithm that adaptively allocates additive noise based on the importance of model parameters. We demonstrate that ANADP narrows the performance gap between regular fine-tuning and traditional DP fine-tuning on a series of datasets while maintaining the required privacy constraints.
When large language models (LLMs) surpass human capabilities, supervising them effectively becomes difficult. Weak-to-strong learning, where a less capable model enhances a stronger one, proves valuable in this context. Yet, the efficacy of this paradigm for complex reasoning tasks is still unexplored. In this paper, we introduce a progressive weak-to-strong reasoning framework that enables the strong model to autonomously refine its training data, maximizing the use of weak signals and unlocking its latent abilities. This framework begins with supervised fine-tuning on a selective small but high-quality dataset, followed by preference optimization on contrastive samples identified by the strong model itself. Experiments on the GSM8K and MATH datasets verify that our method can effectively improve the reasoning capabilities of Llama2-70b using three separate weak models. This work paves the way for a more scalable and sophisticated strategy to enhance AI reasoning powers. All relevant code and resources are available in https://github.com/GAIR-NLP/weak-to-strong-reasoning.
Towards Pareto-Efficient RLHF: Paying Attention to a Few High-Reward Samples with Reward Dropout
PDF ↗Recently, leveraging reinforcement learning (RL) to fine-tune language models (LMs), known as reinforcement learning from human feedback (RLHF), has become an important research topic. However, there is still a lack of theoretical understanding of how RLHF works, the conditions under which it succeeds or fails, and whether it guarantees optimization of both likelihood \beta(\cdot) and reward R(\cdot) objectives. To address these issues, we consider RLHF as a bi-objective problem that has the nature of a Pareto optimization, present a Pareto improvement condition that is necessary to obtain Pareto-efficient policies, and propose a simple yet powerful method named reward dropout that guarantees a Pareto improvement. To demonstrate the performance of reward dropout, two benchmark datasets commonly used in text style transfer tasks were utilized in our study: sentiment and topic datasets sourced from Yelp and AG_News, respectively. Our experiments highlight that paying attention to a few samples with higher rewards leads to greater Pareto improvements regardless of model size. We also demonstrate that the effect of reward dropout is generalizable and most effective with non-pretrained target models, saving the effort of pretraining.
We consider the problem of finding plausible rules that are missing from a given ontology. A number of strategies for this problem have already been considered in the literature. Little is known about the relative performance of these strategies, however, as they have thus far been evaluated on different ontologies. Moreover, existing evaluations have focused on distinguishing held-out ontology rules from randomly corrupted ones, which often makes the task unrealistically easy and leads to the presence of incorrectly labelled negative examples. To address these concerns, we introduce a benchmark with manually annotated hard negatives and use this benchmark to evaluate ontology completion models. In addition to previously proposed models, we test the effectiveness of several approaches that have not yet been considered for this task, including LLMs and simple but effective hybrid strategies.
Large language models (LLMs) have attracted increasing attention due to its prominent performance on various tasks. Recent works seek to leverage LLMs on knowledge-based visual question answering (VQA) tasks which require common sense knowledge to answer the question about an image, since LLMs have obtained rich knowledge from large-scale training. Several methods have proposed to leverage frozen LLMs by converting visual information to textual prompts. However, how to efficiently exploit the knowledge of LLMs and bridge the disconnects between visual information and language models remain open problems. In this paper, we propose to let LLMs learn to ask (L2A) informative questions to collect essential visual information. We introduce the concepts of denotation and connotation to promote image and question understanding and provide a clear guidance with respect to the objective of question generation. In this way, the model can better capture the associations between different concepts, as well as efficiently collect both explicit information and implicit relevant information that contribute to the final answer. The experiments demonstrate that our proposed method achieves consistent performance on various knowledge-based VQA datasets.
There is a variety of multimodal data pertinent to public companies, spanning from accounting statements, macroeconomic statistics, earnings conference calls, and financial reports. These diverse modalities capture the state of firms from a variety of different perspectives but requires complex interactions to reconcile in the formation of accurate financial predictions. The commonality between these different modalities is that they all represent a time series, typically observed for a particular firm at a quarterly horizon, providing the ability to model trends and variations of company data over time. However, the time series of these diverse modalities contains varying temporal and cross-channel patterns that are challenging to model without the appropriate inductive biases. In this work, we design a novel multimodal time series prediction task that includes numerical financial results, macroeconomic states, and long financial documents to predict next quarter’s company earnings relative to analyst expectations. We explore a variety of approaches for this novel setting, establish strong unimodal baselines, and propose a multimodal model that exhibits state-of-the-art performance on this unique task. We demonstrate that each modality contains unique information and that the best performing model requires careful fusion of the different modalities in a multi-stage training approach. To better understand model behavior, we conduct a variety of probing experiments, reveal insights into the value of different modalities, and demonstrate the practical utility of our proposed method in a simulated trading setting.
Is GPT-4V (ision) All You Need for Automating Academic Data Visualization? Exploring Vision-Language Models’ Capability in Reproducing Academic Charts
PDF ↗While effective data visualization is crucial to present complex information in academic research, its creation demands significant expertise in both data management and graphic design. We explore the potential of using Vision-Language Models (VLMs) in automating the creation of data visualizations by generating code templates from existing charts. As the first work to systematically investigate this task, we first introduce AcademiaChart, a dataset comprising 2525 high-resolution data visualization figures with captions from a variety of AI conferences, extracted directly from source codes. We then conduct large-scale experiments with six state-of-the-art (SOTA) VLMs, including both closed-source and open-source models. Our findings reveal that SOTA closed-source VLMs can indeed be helpful in reproducing charts. On the contrary, open-source ones are only effective at reproducing much simpler charts but struggle with more complex ones. Interestingly, the application of Chain-of-Thought (CoT) prompting significantly enhances the performance of the most advanced model, GPT-4-V, while it does not work as well for other models. These results underscore the potential of VLMs in data visualization while also highlighting critical areas that need improvement for broader application.
Although many works have been developed to improve the fairness of deep learning models, their resilience against malicious attacks—particularly the growing threat of backdoor attacks—has not been thoroughly explored.Attacking fairness is crucial because compromised models can introduce biased outcomes, undermining trust and amplifying inequalities in sensitive applications like hiring, healthcare, and law enforcement. This highlights the urgent need to understand how fairness mechanisms can be exploited and to develop defenses that ensure both fairness and robustness. We introduce *BadFair*, a novel backdoored fairness attack methodology. BadFair stealthily crafts a model that operates with accuracy and fairness under regular conditions but, when activated by certain triggers, discriminates and produces incorrect results for specific groups. This type of attack is particularly stealthy and dangerous, as it circumvents existing fairness detection methods, maintaining an appearance of fairness in normal use. Our findings reveal that BadFair achieves a more than 85% attack success rate in attacks aimed at target groups on average while only incurring a minimal accuracy loss. Moreover, it consistently exhibits a significant discrimination score, distinguishing between pre-defined target and non-target attacked groups across various datasets and models.