China

Automatically generated radiology reports often receive high scores from existing evaluation metrics but fail to earn clinicians’ trust. This gap reveals fundamental flaws in how current metrics assess the quality of generated reports. We rethink the design and evaluation of these metrics and propose a clinically grounded Meta-Evaluation Framework. We define clinically grounded criteria spanning clinical alignment and key metric capabilities, including discrimination, robustness, and monotonicity. Using a fine-grained dataset of ground truth and rewritten report pairs annotated with error types and clinical significance labels, we systematically evaluate widely used metrics and uncover their limitations, such as failing to distinguish clinically significant errors, over-penalizing harmless variations, or lacking consistency across error severity levels. Our framework and dataset offer guidance for building more clinically reliable evaluation methods.

EMNLP 2025

ReEvalMed: Rethinking Medical Report Evaluation by Aligning Metrics with Real-World Clinical Judgment

healthcare applications

meta evaluation

clinical nlp

## Welcome!
"I am excited to welcome you to this year’s edition of the Conference on Empirical Methods in Natural Language Processing! Importantly, it marks the 30th edition of EMNLP. With over 8,000 submissions, more than 3,000 accepted papers, and thousands of attendees, we have come a long way from that first
workshop, which had 14 accepted papers. As the field looks ahead, Suzhou is the fitting location for celebrating this milestone: rooted in a long literary tradition, yet modern and forward-looking, and home to a large share of the NLP community."<br>

*Message from the General Chair, Dirk Hovy*





<br>

Celebrate 30 Years of EMNLP! 
EMNLP 2025 will be held in Suzhou, China from November 5th to November 9th, 2025.

poster

Retrieval-augmented Generation (RAG) is powerful, but its effectiveness hinges on which retrievers we use and how. Different retrievers offer distinct, often complementary signals: BM25 captures lexical matches; dense retrievers, semantic similarity. Yet in practice, we typically fix a single retriever based on heuristics, which fails to generalize across diverse information needs. Can we dynamically select and integrate multiple retrievers for each individual query, without the need for manual selection? In our work, we validate this intuition with quantitative analysis and introduce a mixture of retrievers: a zero-shot, weighted combination of heterogeneous retrievers. Extensive experiments show that such mixtures are effective and efficient: Despite totaling just 0.8B parameters, this mixture outperforms every individual retriever and even larger 7B models—by +10.8% and +3.9% on average, respectively. Further analysis also shows that this mixture framework can help incorporate specialized non-oracle human information sources as retrievers to achieve good collaboration, with a 58.9% relative performance improvement over simulated humans alone.

MoR: Better Handling Diverse Queries with a Mixture of Sparse, Dense, and Human Retrievers

Charts are a crucial visual medium for communicating and representing information. While Large Vision-Language Models (LVLMs) have made progress on chart question answering (CQA), the task remains challenging, particularly when models attend to irrelevant regions of the chart. In this work, we present ChartGaze, a new eye-tracking dataset that captures human gaze patterns during chart reasoning tasks. Through a systematic comparison of human and model attention, we find that LVLMs often diverge from human gaze, leading to reduced interpretability and accuracy. To address this, we propose a gaze-guided attention refinement that aligns image-text attention with human fixations. Our approach improves both answer accuracy and attention alignment, yielding gains of up to 2.56 percentage points across multiple models. These results demonstrate the promise of incorporating human gaze to enhance both the reasoning quality and interpretability of chart-focused LVLMs.

ChartGaze: Enhancing Chart Understanding in LVLMs with Eye-Tracking Guided Attention Refinement

Large Multimodal Models (LMMs) exhibit remarkable multi-tasking ability by learning mixed instruction datasets. However, novel tasks would be encountered sequentially in dynamic world, which urges for equipping LMMs with multimodal continual instruction learning (MCIT) ability especially for diverse and challenging generative tasks. However, existing MCIT methods do not fully exploit the unique attribute of LMMs and often gain performance at the expense of efficiency. In this paper, we propose a novel prompt learning framework for MCIT to effectively alleviate forgetting of previous knowledge while managing computational complexity with natural image-text supervision. Concretely, we learn prompts for each task and exploit efficient prompt fusion for knowledge transfer and prompt selection for complexity management with dual-modality guidance. Extensive experiments demonstrate that our approach achieves substantial +14.26% performance gain on MCIT benchmarks with remarkable x1.42 inference speed free from surging computation.

ModalPrompt: Towards Efficient Multimodal Continual Instruction Tuning with Dual-Modality Guided Prompt

Recent advances in multimodal large language models (MLLMs) and diffusion models (DMs) have opened new possibilities for AI-generated content. Yet, personalized cover image generation remains underexplored, despite its critical role in boosting user engagement on digital platforms. We propose ICG, a novel framework that integrates MLLM-based prompting with personalized preference alignment to generate high-quality, contextually relevant covers. ICG extracts semantic features from item titles and reference images via meta tokens, refines them with user embeddings, and injects the resulting personalized context into the diffusion model. To address the lack of labeled supervision, we adopt a multi-reward learning strategy that combines public aesthetic and relevance rewards with a personalized preference model trained from user behavior. Unlike prior pipelines relying on handcrafted prompts and disjointed modules, ICG employs an adapter to bridge MLLMs and diffusion models for end-to-end training. Experiments demonstrate that ICG significantly improves image quality, personalization, user appeal, and downstream recommendation accuracy, providing a scalable solution for real-world content platforms.

ICG: Improving Cover Image Generation via MLLM-based Prompting and Personalized Preference Alignment

Diagnosis based on Electronic Health Records (EHRs) often struggles with data scarcity and privacy concerns. To address these issues, we introduce RareSyn, an innovative data synthesis approach designed to augment and de-identify EHRs, with a focus on rare diseases. The core insight of RareSyn involves using seed EHRs of rare diseases to recall similar records from both common and rare diseases, and then leveraging Large Language Models to substitute the key medical information (e.g., symptoms or examination details) in these records with information from the knowledge graph, thereby generating new EHRs. We first train a transformer Encoder with contrastive learning to integrate various types of medical knowledge. Then, RareSyn engages in iterative processes of recalling similar EHRs, structuring EHRs, revising EHRs, and generating new EHRs until the produced EHRs achieve extensive coverage of the rare disease knowledge. We assess RareSyn based on its utility for diagnosis modeling, the diversity of medical knowledge it incorporates, and the privacy of the synthesized EHRs. Extensive experiments demonstrate its effectiveness in improving disease diagnosis, enhancing diversity, and maintaining privacy.

RareSyn: Health Record Synthesis for Rare Disease Diagnosis

Despite their remarkable progress in multimodal understanding tasks, large vision language models (LVLMs) often suffer from "hallucination'', generating texts misaligned with the visual context. Existing methods aimed at reducing hallucinations through inference time intervention incur a significant increase in latency. To mitigate this, we present **SPIN**, a task-agnostic attention-guided head suppression strategy that can be seamlessly integrated during inference **without incurring any significant compute or latency overhead**. We investigate whether hallucination in LVLMs can be linked to specific model components. Our analysis suggests that hallucinations can be attributed to a dynamic subset of attention heads in each layer. Leveraging this insight, for each text query token, we selectively suppress attention heads that exhibit low attention to image tokens, keeping the top-K attention heads intact. Extensive evaluations on visual question answering and image description tasks demonstrate the efficacy of SPIN in reducing hallucination scores up to **2.7x** while maintaining F1, and improving throughput by **1.8x** compared to existing alternatives.

Mitigating Hallucinations in Vision-Language Models through Image-Guided Head Suppression

Recent advancements in language models have led to significant improvements in mathematical reasoning across various benchmarks. However, most of these benchmarks rely on automatic evaluation methods that only compare final answers using heuristics, without verifying the underlying reasoning steps. This limitation results in false positive solutions, where models may produce correct final answers but with flawed deduction paths. In this paper, we systematically examine the prevalence of false positive solutions in mathematical problem solving for language models. We analyze the characteristics and extent of this issue across different open-source models, datasets of varying difficulty levels, and decoding strategies. Specifically, we explore how false positives influence the inference time scaling behavior of language models. Our experimental results reveal that: (1) false positive solutions persist across different models, datasets, and decoding methods, (2) sampling-based inference time scaling methods do not alleviate the problem, and (3) the pass@N evaluation metric is more susceptible to false positives, suggesting a significantly lower scaling ceiling than what automatic evaluations indicate. Additionally, we analyze specific instances of false positives and discuss potential limitations in self-improvement techniques and synthetic data generation under such conditions.

Examining False Positives under Inference Scaling for Mathematical Reasoning

We introduce ProcWorld, a large-scale benchmark for partially observable embodied spatial reasoning and long-term planning with large language models (LLM) and vision language models (VLM). ProcWorld features a wide range of challenging embodied navigation and object manipulation tasks, covering 16 task types, 5,000 rooms, and over 10 million evaluation trajectories with diverse data distribution. ProcWorld supports configurable observation modes, ranging from text-only descriptions to vision-only observations. It enables text-based actions to control the agent following language instructions. ProcWorld has presented significant challenges for LLMs and VLMs: (1) active information gathering given partial observations for disambiguation; (2) simultaneous localization and decision-making by tracking the spatio-temporal state-action distribution; (3) constrained reasoning with dynamic states subject to physical reachability. Our extensive evaluation of 15 foundation models and 5 reasoning algorithms (with over 1 million rollouts) indicates larger models perform better. However, ProcWorld remains highly challenging for existing state-of-the-art models and in-context learning methods due to constrained reachability and the need for combinatorial spatial reasoning.

ProcWorld: Benchmarking Large Model Planning in Reachability-Constrained Environments

Reasoning is a fundamental capability often required in real-world text-to-image (T2I) generation, e.g., generating "a bitten apple that has been left in the air for more than a week" necessitates understanding temporal decay and commonsense concepts. While recent T2I models have made impressive progress in producing photorealistic images, their reasoning capability remains underdeveloped and insufficiently evaluated. To bridge this gap, we introduce R2I-Bench, a comprehensive benchmark specifically designed to rigorously assess reasoning-driven T2I generation. R2I-Bench comprises 3068 meticulously curated data instances, spanning 7 core reasoning categories, including commonsense, mathematical, logical, compositional, numerical, causal, and concept mixing. To facilitate fine-grained evaluation, we design R2IScore, a QA-style metric based on instance-specific, reasoning-oriented evaluation questions that assess three critical dimensions: text-image alignment, reasoning accuracy, and image quality. Extensive experiments with 16 representative T2I models, including a strong pipeline-based framework that decouples reasoning and generation using the state-of-the-art language and image generation models, demonstrate consistently limited reasoning performance, highlighting the need for more robust, reasoning-aware architectures in the next generation of T2I systems.

R2I-Bench: Benchmarking Reasoning-Driven Text-to-Image Generation

Existing table reasoning methods face challenges due to complex table structures and intricate logical reasoning. While supervised fine-tuning (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 reasoning 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, the Qwen2-VL-7B model with Table-R1 can surpass larger specific tabular reasoning open-source Large Vision Language Models (e.g., 13B Table-LLaVa), even achieving comparable performance to the closed-source model GPT-4o on held-in datasets. Our work highlights the efficacy of each stage of Table-R1 in overcoming initialization bottlenecks and reward sparsity, thereby boosting the GRPO on robust multimodal table reasoning for real-world applications.

Downloads

Next from EMNLP 2025

MoR: Better Handling Diverse Queries with a Mixture of Sparse, Dense, and Human Retrievers

Similar lecture

CLEAR: A Clinically Grounded Tabular Framework for Radiology Report Evaluation