China

We present SportReason, a benchmark for retrieval-augmented reasoning on numerical sports questions. Unlike existing benchmarks limited to one or two evidence units, SportReason requires combining and reasoning across free-text, structured tables, and semi-structured infoboxes. We provide 3,000 human-verified QA pairs by repurposing existing QA and table generation datasets, and by prompting large language models (LLMs). Each pair is grounded in multiple evidence from a multi-modal Wikipedia corpus containing 200K knowledge contexts. We evaluate existing retrievers and rerankers, along with agentic Retrieval-Augmented Generation (RAG) systems. The experimental results show that multi-evidence retrieval remains a challenge. Agentic RAG systems (e.g., Search-o1), despite iterative retrieval and reasoning capabilities, fail to improve performance due to imprecise queries, simple training, and distracting information.

EMNLP 2025

SportReason: Evaluating Retrieval-Augmented Reasoning across Tables and Text for Sports Question Answering

poster

## 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*

[**Link to Conference Handbook**](https://drive.google.com/file/d/1johU5QqVVYO4RfH7QcIORr7qrVBdzdwC/view?usp=sharing)





<br>

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

Fine-tuning large language models (LLMs) often causes overfitting to specific prompt wording, where minor phrasing variations drastically reduce performance. To address this, we propose Prompt-Agnostic Fine-Tuning (PAFT), a method that enhances robustness through dynamic prompt variation during training. PAFT first generates diverse synthetic prompts, then continuously samples from this set to construct training instances, forcing models to learn fundamental task principles rather than surface-level patterns. Across systematic evaluations using both supervised fine-tuning (SFT) and reinforcement learning fine-tuning (RLFT), PAFT consistently demonstrates improved performance on benchmarks for question answering, mathematical reasoning, and tool use. It achieves 7% higher generalization accuracy on unseen prompts than standard methods with similar training efficiency. Notably, models trained with PAFT attain 3.2× faster inference speeds due to reduced prompt sensitivity. Ablation studies further validate effectiveness of PAFT, while theoretical analysis reveals that PAFT can effectively enhance the cross-domain generalization ability of LLM.

PAFT: Prompt-Agnostic Fine-Tuning

The responsible deployment of Large Language Models (LLMs) necessitates rigorous safety evaluation. However, a critical challenge arises from inconsistencies between an LLM's internal refusal decisions and external safety assessments, hindering effective validation. This paper introduces the concept of the 'refusal gap' to define these discrepancies and presents a novel, refusal-aware red teaming framework designed to automatically generate test cases that expose them. Our framework employs refusal probes, potentially leveraging the target model's hidden states, to detect internal model refusals and contrasts these with external safety evaluator judgments. This discrepancy signal then guides a red-teaming model to craft prompts that maximize the refusal gap. To further enhance test case diversity and overcome sparse reward challenges, we introduce a hierarchical curiosity-driven mechanism that rewards both refusal gap maximization and topic exploration. Empirical results demonstrate that our method significantly outperforms existing RL-based approaches in generating diverse test cases and achieves a substantially higher discovery rate of refusal gaps.

Refusal-Aware Red Teaming: Exposing Inconsistency in Safety Evaluations

Existing large language model (LLM) agents for automating data science show promise, but they remain constrained by narrow task scopes, limited generalization across tasks and models, and over-reliance on state-of-the-art (SOTA) LLMs. We introduce DatawiseAgent, a notebook-centric LLM agent framework for adaptive and robust data science automation. Inspired by how human data scientists work in computational notebooks, DatawiseAgent introduces a unified interaction representation and a multi-stage architecture based on finite-state transducers (FSTs). This design enables flexible long-horizon planning, progressive solution development, and robust recovery from execution failures. Extensive experiments across diverse data science scenarios and models show that DatawiseAgent consistently achieves SOTA performance by surpassing strong baselines such as AutoGen and TaskWeaver, demonstrating superior effectiveness and adaptability. Further evaluations reveal graceful performance degradation under weaker or smaller models, underscoring the robustness and scalability.

DatawiseAgent: A Notebook-Centric LLM Agent Framework for Adaptive and Robust Data Science Automation

Reasoning large language models (LLMs) excel in complex tasks, which has drawn significant attention to reinforcement learning (RL) for LLMs. However, existing approaches allocate an equal number of rollouts to all questions during the RL process, which is inefficient. This inefficiency stems from the fact that training on simple questions yields limited gains, whereas more rollouts are needed for challenging questions to sample correct answers. Furthermore, while RL improves response precision, it limits the model's exploration ability, potentially resulting in a performance cap below that of the base model prior to RL. To address these issues, we propose a mechanism for dynamically allocating rollout budgets based on the difficulty of the problems, enabling more efficient RL training. Additionally, we introduce an adaptive dynamic temperature adjustment strategy to maintain the entropy at a stable level, thereby encouraging sufficient exploration. This enables LLMs to improve response precision while preserving their exploratory ability to uncover potential correct pathways. The code and data is available on: https://anonymous.4open.science/r/E3-RL4LLMs-DB28

Enhancing Efficiency and Exploration in Reinforcement Learning for LLMs

Existing large language model (LLM) evaluation benchmarks primarily focus on English, while current multilingual tasks lack parallel questions that specifically assess cross-linguistic reasoning abilities. This dual limitation makes it challenging to comprehensively assess LLMs' performance in the multilingual setting. To fill this gap, we introduce MMLU-ProX, a comprehensive benchmark covering 29 languages, built on an English benchmark. Each language version consists of 11,829 identical questions, enabling direct cross-linguistic comparisons. Additionally, to meet efficient evaluation needs, we provide a lite version containing 658 questions per language. To ensure the high quality of MMLU-ProX, we employ a rigorous development process that involves multiple powerful LLMs for translation, followed by expert review to ensure accurate expression, consistent terminology, and cultural relevance. Building on this, we systematically evaluate 36 state-of-the-art LLMs, including reasoning-enhanced and multilingual-optimized LLMs. The results reveal significant disparities in the multilingual capabilities of LLMs: While they perform well in high-resource languages, their performance declines markedly in low-resource languages, with gaps of up to 24.3%. Through MMLU-ProX, we aim to advance the development of more inclusive AI systems and promote equitable access to technology across global contexts.

MMLU-ProX: A Multilingual Benchmark for Advanced Large Language Model Evaluation

Large language models (LLMs) have emerged as a potential solution to automate the complex processes involved in writing literature reviews, such as literature collection, organization, and summarization. However, it is yet unclear how good LLMs are at automating comprehensive and reliable literature reviews. This study introduces a framework to automatically evaluate the performance of LLMs in three key tasks of literature review writing: reference generation, abstract writing, and literature review composition. We introduce multidimensional evaluation metrics that assess the hallucination rates in generated references and measure the semantic coverage and factual consistency of the literature summaries and compositions against human-written counterparts. The experimental results reveal that even the most advanced models still generate hallucinated references, despite recent progress. Moreover, we observe that the performance of different models varies across disciplines when it comes to writing literature reviews. These findings highlight the need for further research and development to improve the reliability of LLMs in automating academic literature reviews. The dataset and code used in this study are publicly available at an anonymous repository~\footnote{\url{https://anonymous.4open.science/r/Eval_LLM_LR-C657}}

Large Language Models for Automated Literature Review: An Evaluation of Reference Generation, Abstract Writing, and Review Composition

Thanks to their linguistic capabilities, LLMs offer an opportunity to bridge the gap between informal mathematics and formal languages through autoformalization. However, it is still unclear how well LLMs generalize to sophisticated and naturally occurring mathematical statements. To address this gap, we investigate the task of autoformalizing real-world mathematical definitions -- a critical component of mathematical discourse. Specifically, we introduce two novel resources for autoformalization, collecting definitions from Wikipedia (Def_Wiki) and arXiv papers (Def_ArXiv). We then systematically evaluate a range of LLMs, analyzing their ability to formalize definitions into Isabelle/HOL. Furthermore, we investigate strategies to enhance LLMs' performance including refinement through external feedback from Proof Assistants, and formal definition grounding, where we augment LLMs' formalizations through relevant contextual elements from formal mathematical libraries. Our findings reveal that definitions present a greater challenge compared to existing benchmarks, such as miniF2F. In particular, we found that LLMs still struggle with self-correction, and aligning with relevant mathematical libraries. At the same time, structured refinement methods and definition grounding strategies yield notable improvements of up to 16% on self-correction capabilities and 43% on the reduction of undefined errors, highlighting promising directions for enhancing LLM-based autoformalization in real-world scenarios.

Autoformalization in the Wild: Assessing LLMs on Real-World Mathematical Definitions

The rapid progress of Large Language Models has advanced agentic systems in decision-making, coordination, and task execution. Yet, existing agentic system generation frameworks lack full autonomy, missing from-scratch agent generation, self-optimizing functionality, and collaboration, limiting adaptability and automation. We propose **SwarmAgentic**, a framework for fully automated agentic system generation, extending Particle Swarm Optimization (PSO) into a language-driven search space for structure-level optimization. SwarmAgentic instantiates agents from scratch and jointly optimizes agent functionality and collaboration as interdependent components. We evaluate our method on six real-world, open-ended, and exploratory tasks involving high-level planning, system-level coordination, and creative reasoning. Given only a task description and an objective function, SwarmAgentic outperforms all baselines, achieving a **+261.8% relative improvement** over ADAS on the TravelPlanner benchmark, highlighting the effectiveness of full automation in structurally unconstrained tasks. This framework marks a significant step toward scalable and autonomous agentic system design, bridging swarm intelligence with fully automated system multi-agent generation.

SwarmAgentic: Towards Fully Automated Agentic System Generation via Swarm Intelligence

This paper introduces TurnaboutLLM, a novel framework and dataset for evaluating the deductive reasoning abilities of Large Language Models (LLMs) by leveraging the interactive gameplay of detective games Ace Attorney and Danganronpa. The framework tasks LLMs with identifying contradictions between testimonies and evidences within long narrative contexts, a challenging task due to the large answer space and diverse reasoning types presented by its questions. We evaluate twelve state-of-the-art LLMs on the dataset, hinting at limitations of popular strategies for enhancing deductive reasoning such as extensive thinking and Chain-of-Thought prompting. The results also suggest varying effects of context size, reasoning steps and answer space size on model performance. Overall, TurnaboutLLM presents a substantial challenge for LLMs' deductive reasoning abilities in complex, narrative-rich environments.

TurnaboutLLM: A Deductive Reasoning Benchmark from Detective Games

Construction grammar posits that constructions, or form-meaning pairings, are acquired through experience with language (the distributional learning hypothesis). But how much information about constructions does this distribution actually contain? Corpus-based analyses provide some answers, but text alone cannot answer counterfactual questions about what caused a particular word to occur. For that, we need computable models of the distribution over strings---namely, pretrained language models (PLMs). Here we treat a RoBERTa model as a proxy for this distribution and hypothesize that constructions will be revealed within it as patterns of statistical affinity. We support this hypothesis experimentally: many constructions are robustly distinguished, including (i) hard cases where semantically distinct constructions are superficially similar, as well as (ii) schematic constructions, whose ``slots'' can be filled by abstract word classes. Despite this success, we also provide qualitative evidence that statistical affinity alone may be insufficient to identify all constructions from text. Thus, statistical affinity is likely an important, but partial, signal available to learners.

Downloads

Next from EMNLP 2025

PAFT: Prompt-Agnostic Fine-Tuning

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES