Zimu Wang

With the advancement of large language models (LLMs) and multimodal LLMs (MLLMs), table reasoning has achieved significant progress. However, most existing works focus predominantly on textual or rendered tables, which differ substantially from real-world photographed scenarios under suboptimal conditions such as uneven lighting, blur, and tilted perspectives. This discrepancy creates a significant performance gap for current MLLMs, limiting their applicability in real-world scenarios. To address this critical limitation, we introduce the first comprehensive study on multimodal reasoning for photographed tables. We present a new dataset Photographed Chinese Table Reasoning (PCTR-16K), which contains 4989 photographed tables and 16,318 questions across 9 subjects and 3 difficulty levels. This dataset serves as the first benchmark specifically designed for evaluating reasoning capabilities on tables captured under authentic conditions. To enhance MLLMs’ reasoning over photographed tables, we propose Structure-aware Chain-of-Thought (SCoT), a method that unifies table recognition and reasoning into a single end-to-end generative process. To bolster the structural perception required by SCoT, we further incorporate seven auxiliary table structure understanding (TSU) tasks during fine-tuning. These tasks provide fine-grained supervision across multiple dimensions of table layout and semantics. Extensive experiments on various MLLMs demonstrate that our proposed SCoT and multi-view TSU tasks significantly enhance recognition and reasoning capabilities on photographed tables. For example, LLaVA-Llama3.1 achieves an absolute improvement of 19.53% on the PCTR-16K benchmark (from 44.58% to 64.11%), demonstrating its effectiveness for real-world photographed table reasoning. The dataset is publicly available at https://github.com/PremiLab-Math/PCTR-16k.

Psychological defenses are strategies, often automatic, that people use to manage distress. Rigid or overuse of defenses is negatively linked to mental health and shapes what speakers disclose and how they accept or resist help. However, defenses are complex and difficult to reliably measure, particularly in clinical dialogues. We introduce PsyDefConv, a dialogue corpus with help seeker utterances labeled for defense level, and DMRS Co-Pilot, a four-stage pipeline that provides evidence-based pre-annotations. The corpus contains 200 dialogues and 4709 utterances, including 2336 help seeker turns, with labeling and Cohen’s kappa 0.639. In a counterbalanced study, the co-pilot reduced average annotation time by 22.4%. In expert review, it averaged 4.62 for evidence, 4.44 for clinical plausibility, and 4.40 for insight on a seven-point scale. Benchmarks with strong language models in zero-shot and fine-tuning settings demonstrate clear headroom, with the best macro F1-score around 30% and a tendency to overpredict mature defenses. Corpus analyses confirm that mature defenses are most common and reveal emotion-specific deviations. We will release the corpus, annotations, code, and prompts to support research on defensive functioning in language.

The increasing integration of large language models (LLMs) into mental health applications necessitates robust frameworks for evaluating professional safety alignment. Current evaluative approaches primarily rely on refusal-based safety signals, which offer limited insight into the nuanced behaviors required in clinical practice. In mental health, clinically inadequate refusals can be perceived as unempathetic and discourage help-seeking. To address this gap, we move beyond refusal-centric metrics and introduce PsychEthicsBench, the first principle-grounded benchmark based on Australian psychology and psychiatry guidelines, designed to evaluate LLMs’ ethical knowledge and behavioral responses through multiple-choice and open-ended tasks with fine-grained ethicality annotations. Empirical results across 14 models reveal that refusal rates are poor indicators of ethical behavior, revealing a significant divergence between safety triggers and clinical appropriateness. Notably, we find that domain-specific fine-tuning can degrade ethical robustness, as several specialized models underperform their base backbones in ethical alignment. PsychEthicsBench provides a foundation for systematic, jurisdiction-aware evaluation of LLMs in mental health, encouraging more responsible development in this domain.

As AI agents improve, the central question is no longer whether they can solve isolated well-defined financial tasks, but whether they can reliably carry out financial professional work. Existing financial benchmarks offer only a partial view of this ability, as they primarily evaluate static competencies such as question answering, retrieval, summarization, and classification. We introduce Herculean, the first skilled benchmark for agentic financial intelligence spanning four representative workflows, including Trading, Hedging, Market Insights, and Auditing. Each workflow is instantiated as a standardized MCP-based skill environment with its own tools, interaction dynamics, constraints, and success criteria, enabling consistent end-to-end assessment of heterogeneous agent systems. Across frontier agents, we find agents perform relatively well on Trading and Market Insights, but struggle substantially on Hedging and Auditing, where long-horizon coordination, state consistency, and structured verification are critical. Overall, our results point to a key gap in current agents in turning financial reasoning into dependable workflow execution in high-stakes financial workflows.

Graph-based retrieval-augmented generation (GraphRAG) has recently emerged as a powerful paradigm for knowledge-intensive question answering, especially for tasks that require structured evidence organization and multi-hop reasoning. However, existing GraphRAG systems are typically built in a one-size-fits-all manner, relying on a fixed retrieval framework and a single, often large and costly, generator LLM for all queries. This static design limits their ability to adapt to the complexity of varying questions and often incurs unnecessary computational cost. To fill in the gap, we propose GraphRAG-Router, a cost-efficient framework that adopts a hierarchical routing strategy to coordinate heterogeneous GraphRAGs and generator LLMs. Specifically, GraphRAG-Router is first warmed up through supervised fine-tuning and then optimized with a two-stage reinforcement learning procedure, whose second stage introduces a curriculum cost-aware reward to encourage difficulty-aware and economical generator allocation. Extensive experiments on six general-domain and multi-hop QA benchmarks show that GraphRAG-Router consistently outperforms state-of-the-art baselines, reducing the overuse of large LLMs by nearly 30% while maintaining strong generalization capability.

The increasing use of large language models (LLMs) in mental healthcare raises safety concerns in high-stakes therapeutic interactions. A key challenge is distinguishing therapeutic empathy from maladaptive validation, where supportive responses may inadvertently reinforce harmful beliefs or behaviors in multi-turn conversations. This risk is largely overlooked by existing red-teaming frameworks, which focus mainly on generic harms or optimization-based attacks. To address this gap, we introduce Personality-based Client Simulation Attack (PCSA), the first red-teaming framework that simulates clients in psychological counseling through coherent, persona-driven client dialogues to expose vulnerabilities in psychological safety alignment. Experiments on seven general and mental health-specialized LLMs show that PCSA substantially outperforms four competitive baselines. Perplexity analysis and human inspection further indicate that PCSA generates more natural and realistic dialogues. Our results reveal that current LLMs remain vulnerable to domain-specific adversarial tactics, providing unauthorized medical advice, reinforcing delusions, and implicitly encouraging risky actions.

The medical adoption of NLP tools requires interpretability by end users, yet traditional explainable AI (XAI) methods are misaligned with clinical reasoning and lack clinician input. We introduce CHiRPE (Clinical High-Risk Prediction with Explainability), an NLP pipeline that takes transcribed semi-structured clinical interviews to: (i) predict psychosis risk; and (ii) generate novel SHAP explanation formats co-developed with clinicians. Trained on 944 semi-structured interview transcripts across 24 international clinics of the AMP-SCZ study, the CHiRPE pipeline integrates symptom-domain mapping, LLM summarisation, and BERT classification. CHiRPE achieved over 90% accuracy across three BERT variants and outperformed baseline models. Explanation formats were evaluated by 28 clinical experts who indicated a strong preference for our novel concept-guided explanations, especially hybrid graph-and-text summary formats. CHiRPE demonstrates that clinically-guided model development produces both accurate and interpretable results. Our next step is focused on real-world testing across our 24 international sites.

Depression is one of the most prevalent mental health disorders globally. In recent years, multi-modal data, such as speech, video, and transcripts, has been increasingly used to develop AI-assisted depression assessment systems. Large language models have further advanced this field due to their strong language understanding and generalization capabilities. However, conventional LLMs remain text-centric and cannot process the rich non-verbal cues found in audio and visual modalities, which are critical components in mental health evaluation. While multi-modal LLMs offer a promising direction, few are tailored for psychological applications. In this study, we propose a novel multi-modal LLM framework for depression detection. Our approach augments an audio language model with visual understanding and aligns audio-visual features at the timestamp level. This fine-grained alignment improves modeling of temporal dynamics across modalities while reducing the need for extensive training data and computational resources. Experiments on the DAIC-WoZ dataset demonstrate that our model outperforms both single-modality approaches and previous multi-modal methods. Moreover, the proposed framework can be extended to incorporate additional physiological signals, paving the way for broader clinical applications beyond mental health.

Large language model (LLM) agents have demonstrated remarkable potential in advancing scientific discovery. However, their capability in the fundamental yet crucial task of reproducing code from research papers, especially in the NLP domain, remains underexplored. This task includes unique complex reasoning challenges in the intellectual synthesis of abstract concepts and the comprehension of code repositories with interdependent files. Motivated by this gap, we present LMR-BENCH, a benchmark designed to systematically evaluate the capability of LLM agents on code reproduction from Language Modeling Research. It consists of 28 code reproduction tasks derived from 23 research papers published in top-tier NLP venues over the past five years, spanning nine fundamental categories. Models are provided with a research paper, a code repository containing one or more masked functions, and instructions for implementing these functions. We conduct extensive experiments in standard prompting and LLM agent settings with state-of-the-art LLMs, evaluating the accuracy of unit tests and performing LLM-based evaluation of code correctness. Experimental results reveal that even the most advanced models still exhibit persistent limitations in scientific reasoning and code synthesis, highlighting critical gaps in LLM agents’ ability to autonomously reproduce scientific research.

Medical fact-checking has become increasingly critical as more individuals seek medical information online. However, existing datasets predominantly focus on human-generated content, leaving the verification of content generated by large language models (LLMs) relatively unexplored. To address this gap, we introduce MedFact, the first evidence-based Chinese medical fact-checking dataset of LLM-generated medical content. It consists of 1,321 questions and 7,409 claims, mirroring the complexities of real-world medical scenarios. We conduct comprehensive experiments in both in-context learning (ICL) and fine-tuning settings, showcasing the capability and challenges of current LLMs on this task, accompanied by an in-depth error analysis to point out key directions for future research. Our dataset is publicly available at https://github.com/AshleyChenNLP/MedFact.

Existing table understanding methods face challenges due to complex table structures and intricate logical reasoning. While supervised finetuning (SFT) dominates existing research, reinforcement learning (RL), such as Group Relative Policy Optimization (GRPO), has shown promise but struggled with low initial policy accuracy and coarse rewards in tabular contexts. In this paper, we introduce Table-R1, a three-stage RL framework that enhances multimodal table understanding through: (1) Warm-up that prompts initial perception and reasoning capabilities, (2) Perception Alignment GRPO (PA-GRPO), which employs continuous Tree-Edit-Distance Similarity (TEDS) rewards for recognizing table structures and contents, and (3) Hint-Completion GRPO (HC-GRPO), which utilizes fine-grained rewards of residual steps based on the hint-guided question. Extensive experiments demonstrate that Table-R1 can boost the model’s table reasoning performance obviously on both held-in and held-out datasets, outperforming SFT and GRPO largely. Notably, Qwen2-VL-7B with Table-R1 surpasses larger specific table understanding models (e.g., Table-LLaVA 13B), even achieving comparable performance to the closed-source model GPT-4o on held-in datasets, demonstrating the efficacy of each stage of Table-R1 in overcoming initialization bottlenecks and reward sparsity, thereby advancing robust multimodal table understanding.

Multimodal Large Language Models (MLLMs) have shown promise in visual-textual reasoning, with Multimodal Chain-of-Thought (MCoT) prompting significantly enhancing interpretability. However, existing MCoT methods rely on rationale-rich datasets and largely focus on inter-object reasoning, overlooking the intra-object understanding crucial for image classification. To address this gap, we propose WISE, a Weak-supervision-guided Step-by-step Explanation method that augments any image classification dataset with MCoTs by reformulating the concept-based representations from Concept Bottleneck Models (CBMs) into concise, interpretable reasoning chains under weak supervision. Experiments across ten datasets show that our generated MCoTs not only improve interpretability by 37% but also lead to gains in classification accuracy when used to fine-tune MLLMs. Our work bridges concept-based interpretability and generative MCoT reasoning, providing a generalizable framework for enhancing MLLMs in fine-grained visual understanding.

Phonetic Cloaking Replacement (PCR), defined as the deliberate use of homophonic or near-homophonic variants to hide toxic intent, has become a major obstacle to Chinese content moderation. While this problem is well-recognized, existing evaluations predominantly rely on rule-based, synthetic perturbations that ignore the creativity of real users. We organize PCR into a four-way surface-form taxonomy and compile PCR-ToxiCN, a dataset of 500 naturally occurring, phonetically cloaked offensive posts gathered from the RedNote platform. Benchmarking state-of-the-art LLMs on this dataset exposes a serious weakness: the best model reaches only an F1-score of 0.672, and zero-shot chain-of-thought prompting pushes performance even lower. Guided by error analysis, we revisit a Pinyin-based prompting strategy that earlier studies judged ineffective and show that it recovers much of the lost accuracy. This study offers the first comprehensive taxonomy of Chinese PCR, a realistic benchmark that reveals current detectors’ limits, and a lightweight mitigation technique that advances research on robust toxicity detection.

Recent advancements in 2D multimodal large language models (MLLMs) have significantly improved performance in vision-language tasks. However, extending these capabilities to 3D environments remains a distinct challenge due to the complexity of spatial reasoning. Nevertheless, existing 3D benchmarks often lack fine-grained numerical reasoning task annotations, limiting MLLMs’ ability to perform precise spatial measurements and complex numerical reasoning. To address this gap, we introduce NUMINA, the first Natural Understanding benchmark for Multi-dimensional Intelligence and Numerical reasoning Abilities to enhance multimodal indoor perceptual understanding. NUMINA features multi-scale annotations and various question-answer pairs, generated using NUMINA-Flow, an automated annotation pipeline that integrates LLM rewriting and rule-based self-verification. We evaluate the performance of various state-of-the-art LLMs on NUMINA following the Chat-Scene framework, demonstrating that current LLMs struggle with multimodal numerical reasoning, particularly in performing precise computations such as distance and volume estimation, highlighting the need for further advancements in 3D models. The dataset and source codes can be obtained from https://github.com/fengshun124/NUMINA

Transformer-based deep neural networks have achieved remarkable success across various computer vision tasks, largely attributed to their long-range self-attention mechanism and scalability. However, most transformer architectures embed images into uniform, grid-based vision tokens, neglecting the underlying semantic meanings of image regions, resulting in suboptimal feature representations. To address this issue, we propose Fuzzy Token Clustering Transformer (FTCFormer), which incorporates a novel clustering-based downsampling module to dynamically generate vision tokens based on the semantic meanings instead of spatial positions. It allocates fewer tokens to less informative regions and more to represent semantically important regions, regardless of their spatial adjacency or shape irregularity. To further enhance feature extraction and representation, we propose a Density Peak Clustering-Fuzzy K-Nearest Neighbor (DPC-FKNN) mechanism for clustering center determination, a Spatial Connectivity Score (SCS) for token assignment, and a channel-wise merging (Cmerge) strategy for token merging. Extensive experiments on 32 datasets across diverse domains validate the effectiveness of FTCFormer on image classification, showing consistent improvements over the TCFormer baseline, achieving gains of improving 1.43% on five fine-grained datasets, 1.09% on six natural image datasets, 0.97% on three medical datasets and 0.55% on four remote sensing datasets. The code is available at: https://github.com/BaoBao0926/FTCFormer/tree/main.

As large language models (LLMs) integrate into collaborative teams, their social conformity – the tendency to align with majority opinions – has emerged as a key concern. In humans, conformity arises from informational influence (rational use of group cues for accuracy) or normative influence (social pressure for approval), with uncertainty moderating this balance by shifting from purely analytical to heuristic processing. It remains unclear whether these human psychological mechanisms apply to LLMs. This study adapts the information cascade paradigm from behavioral economics to quantitatively disentangle the two drivers to investigate the moderate effect. We evaluated nine leading LLMs across three decision-making scenarios (medical, legal, investment), manipulating information uncertainty (q = 0.667, 0.55, and 0.70, respectively). Our results indicate that informational influence underpins the models’ behavior across all contexts, with accuracy and confidence consistently rising with stronger evidence. However, this foundational mechanism is dramatically modulated by uncertainty. In low-to-medium uncertainty scenarios, this informational process is expressed as a conservative strategy, where LLMs systematically underweight all evidence sources. In contrast, high uncertainty triggers a critical shift: while still processing information, the models additionally exhibit a normative-like amplification, causing them to overweight public signals (beta > 1.55 vs. private beta = 0.81).

Mental health remains a critical global challenge, with increasing demand for accessible, effective interventions. Large language models (LLMs) offer promising solutions in psychotherapy by enhancing the assessment, diagnosis, and treatment of mental health conditions through dynamic, context-aware interactions. This survey provides a comprehensive overview of the current landscape of LLM applications in psychotherapy, highlighting the roles of LLMs in symptom detection, severity estimation, cognitive assessment, and therapeutic interventions. We present a novel conceptual taxonomy to organize the psychotherapy process into three core components: assessment, diagnosis, and treatment, and examine the challenges and advancements in each area. The survey also addresses key research gaps, including linguistic biases, limited disorder coverage, and underrepresented therapeutic models. Finally, we discuss future directions to integrate LLMs into a holistic, end-to-end psychotherapy framework, addressing the evolving nature of mental health conditions and fostering more inclusive, personalized care.

Medical image segmentation is prerequisite in computer-aided diagnosis. As the field experiences tremendous paradigm changes since the introduction of foundation models, technicality of deep medical segmentation model is no longer a privilege limited to computer science researchers. A comprehensive educational resource suitable for researchers of broad, different backgrounds such as biomedical and medicine, is needed. This review strategically covers the evolving trends that happens to different fundamental components of medical image segmentation such as the emerging of multimodal medical image datasets, updates on deep learning libraries, classical-to-contemporary development in deep segmentation models and latest challenges with focus on enhancing the interpretability and generalizability of model. Last, the conclusion section highlights on future trends in deep medical segmentation that worth further attention and investigations.

Solving tabular math word problems (TMWPs) has become a critical role in evaluating the mathematical reasoning ability of large language models (LLMs), where large-scale TMWP samples are commonly required for LLM fine-tuning. Since the collection of high-quality TMWP datasets is costly and time-consuming, recent research has concentrated on automatic TMWP generation. However, current generated samples usually suffer from issues of either correctness or diversity. In this paper, we propose a Template-driven LLM-paraphrased (TeLL) framework for generating high-quality TMWP samples with diverse backgrounds and accurate tables, questions, answers, and solutions. To this end, we first extract templates from existing real samples to generate initial problems, ensuring correctness. Then, we adopt an LLM to extend templates and paraphrase problems, obtaining diverse TMWP samples. Furthermore, we find the reasoning annotation is important for solving TMWPs. Therefore, we propose to enrich each solution with illustrative reasoning steps. Through the proposed framework, we construct a high-quality dataset TabMWP-TeLL by adhering to the question types in the TabMWP dataset, and we conduct extensive experiments on a variety of LLMs to demonstrate the effectiveness of TabMWP-TeLL in improving TMWP solving performance. The code and data of this paper are available at: https://github.com/Jason8Kang/TELL.

Mental manipulation severely undermines mental wellness by covertly and negatively distorting decision-making. While there is an increasing interest in mental health care within the natural language processing community, progress in tackling manipulation remains limited due to the complexity of detecting subtle, covert tactics in conversations. In this paper, we propose Intent-Aware Prompting (IAP), a novel approach for detecting mental manipulations using large language models (LLMs), providing a deeper understanding of manipulative tactics by capturing the underlying intents of participants. Experimental results on the MentalManip dataset demonstrate superior effectiveness of IAP against other advanced prompting strategies. Notably, our approach substantially reduces false negatives, helping detect more instances of mental manipulation with minimal misjudgment of positive cases. The code of this paper is available at https://github.com/Anton-Jiayuan-MA/Manip-IAP.

Existing benchmarks for multi-hop question answering (QA) primarily evaluate models based on their ability to reason about entities and the relationships between them. However, there’s a lack of insight into how these models perform in terms of both events and entities. In this paper, we introduce a novel semi-automatic question generation strategy by composing event structures from information extraction (IE) datasets and present the first Multi-hop Event-centric Question Answering (MEQA) benchmark. It contains (1) 2,243 challenging questions that require a diverse range of complex reasoning over entity-entity, entity-event, and event-event relations; (2) corresponding multi-step QA-format event reasoning chain (explanation) which leads to the answer for each question. We also introduce two metrics for evaluating explanations: completeness and logical consistency. We conduct comprehensive benchmarking and analysis, which shows that MEQA is challenging for the latest state-of-the-art models encompassing large language models (LLMs); and how they fall short of providing faithful explanations of the event-centric reasoning process.

As an essential task in information extraction (IE), Event-Event Causal Relation Extraction (ECRE) aims to identify and classify the causal relationships between event mentions in natural language texts. However, existing research on ECRE has highlighted two critical challenges, including the lack of document-level modeling and causal hallucinations. In this paper, we propose a Knowledge-guided binary Question Answering (KnowQA) method with event structures for ECRE, consisting of two stages: Event Structure Construction and Binary Question Answering. We conduct extensive experiments under both zero-shot and fine-tuning settings with large language models (LLMs) on the MECI and MAVEN-ERE datasets. Experimental results demonstrate the usefulness of event structures on document-level ECRE and the effectiveness of KnowQA by achieving state-of-the-art on the MECI dataset. We observe not only the effectiveness but also the high generalizability and low inconsistency of our method, particularly when with complete event structures after fine-tuning the models.

Deep learning-based natural language processing (NLP) models, particularly pre-trained language models (PLMs), have been revealed to be vulnerable to adversarial attacks. However, the adversarial examples generated by many mainstream word-level adversarial attack models are neither valid nor natural, leading to the loss of semantic maintenance, grammaticality, and human imperceptibility. Based on the exceptional capacity of language understanding and generation of large language models (LLMs), we propose LLM-Attack, which aims at generating both valid and natural adversarial examples with LLMs. The method consists of two stages: word importance ranking (which searches for the most vulnerable words) and word synonym replacement (which substitutes them with their synonyms obtained from LLMs). Experimental results on the Movie Review (MR), IMDB, and Yelp Review Polarity datasets against the baseline adversarial attack models illustrate the effectiveness of LLM-Attack, and it outperforms the baselines in human and GPT-4 evaluation by a significant margin. The model can generate adversarial examples that are typically valid and natural, with the preservation of semantic meaning, grammaticality, and human imperceptibility.

The goal of portfolio management is to simultaneously maximize the accumulated return and also to control risk. In consecutive trading periods, portfolio manager needs to continuously adjust the portfolio weights based on the factors which can cause price fluctuation in the market. In the stock market, the factors affecting the stock price can be divided into two categories. The first is price fluctuations caused by irrational investment of the speculators. The second is endogenous value changes caused by operations of the company. In recent years, with the advancement of artificial intelligence technology, reinforcement learning (RL) algorithms have been increasingly employed by scholars to address financial problems, particularly in the area of portfolio management. However, the deep RL models proposed by these scholars in the past have focused more on analyzing the price changes caused by the investment behavior of speculators in response to technical indicators of actual stock prices. In this research, we introduce an RL-based framework called FinBPM, which takes both the factor pertaining to the impact on operations of the company and the factor of the irrational investment of the speculator into consideration. For our experimentation, we randomly selected twelve stocks from the Dow Jones Industrial Index to construct our portfolio. The experimental results reveal that, in comparison to conventional reinforcement learning methods, our approach with at least 13.26% increase over other methods compared. Additionally, it achieved the best Sharpe ratio of 2.77, effectively maximizing the return per unit of risk.

Event understanding aims at understanding the content and relationship of events within texts, which covers multiple complicated information extraction tasks: event detection, event argument extraction, and event relation extraction. To facilitate related research and application, we present an event understanding toolkit OmniEvent, which features three desiderata: (1) Comprehensive. OmniEvent supports mainstream modeling paradigms of all the event understanding tasks and the processing of 15 widely-used English and Chinese datasets. (2) Fair. OmniEvent carefully handles the inconspicuous evaluation pitfalls reported in Peng et al. (2023), which ensures fair comparisons between different models. (3) Easy-to-use. OmniEvent is designed to be easily used by users with varying needs. We provide off-the-shelf models that can be directly deployed as web services. The modular framework also enables users to easily implement and evaluate new event understanding models with OmniEvent. The toolkit is publicly released along with the demonstration website and video.

In-context learning (ICL) has become the default method for using large language models (LLMs), making the exploration of its limitations and understanding the underlying causes crucial. In this paper, we find that ICL falls short of handling specification-heavy tasks, which are tasks with complicated and extensive task specifications, requiring several hours for ordinary humans to master, such as traditional information extraction tasks. The performance of ICL on these tasks mostly cannot reach half of the state-of-the-art results. To explore the reasons behind this failure, we conduct comprehensive experiments on 18 specification-heavy tasks with various LLMs and identify three primary reasons: inability to specifically understand context, misalignment in task schema comprehension with humans, and inadequate long-text understanding ability. Furthermore, we demonstrate that through fine-tuning, LLMs can achieve decent performance on these tasks, indicating that the failure of ICL is not an inherent flaw of LLMs, but rather a drawback of existing alignment methods that renders LLMs incapable of handling complicated specification-heavy tasks via ICL. To substantiate this, we perform dedicated instruction tuning on LLMs for these tasks and observe a notable improvement. We hope the analyses in this paper could facilitate advancements in alignment methods enabling LLMs to meet more sophisticated human demands.

The diverse relationships among real-world events, including coreference, temporal, causal, and subevent relations, are fundamental to understanding natural languages. However, two drawbacks of existing datasets limit event relation extraction (ERE) tasks: (1) Small scale. Due to the annotation complexity, the data scale of existing datasets is limited, which cannot well train and evaluate data-hungry models. (2) Absence of unified annotation. Different types of event relations naturally interact with each other, but existing datasets only cover limited relation types at once, which prevents models from taking full advantage of relation interactions. To address these issues, we construct a unified large-scale human-annotated ERE dataset MAVEN-ERE with improved annotation schemes. It contains 103,193 event coreference chains, 1,216,217 temporal relations, 57,992 causal relations, and 15,841 subevent relations, which is larger than existing datasets of all the ERE tasks by at least an order of magnitude. Experiments show that ERE on MAVEN-ERE is quite challenging, and considering relation interactions with joint learning can improve performances. The dataset and source codes can be obtained from https://github.com/THU-KEG/MAVEN-ERE.