China

We present a novel, open-source social network simulation framework, MOSAIC, where generative language agents predict user behaviors such as liking, sharing, and flagging content. This simulation combines LLM agents with a directed social graph to analyze emergent deception behaviors and gain a better understanding of how users determine the veracity of online social content. By constructing user representations from diverse fine-grained personas, our system enables multi-agent simulations that model content dissemination and engagement dynamics at scale. Within this framework, we evaluate three different content moderation strategies with simulated misinformation dissemination, and we find that they not only mitigate the spread of non-factual content but also increase user engagement. In addition, we analyze the trajectories of popular content in our simulations, and explore whether simulation agents&#39; articulated reasoning for their social interactions truly aligns with their collective engagement patterns.

EMNLP 2025

MOSAIC: Modeling Social AI for Content Dissemination and Regulation in Multi-Agent Simulations

We present a novel, open-source social network simulation framework, MOSAIC, where generative language agents predict user behaviors such as liking, sharing, and flagging content. This simulation combines LLM agents with a directed social graph to analyze emergent deception behaviors and gain a better understanding of how users determine the veracity of online social content. By constructing user representations from diverse fine-grained personas, our system enables multi-agent simulations that model content dissemination and engagement dynamics at scale. Within this framework, we evaluate three different content moderation strategies with simulated misinformation dissemination, and we find that they not only mitigate the spread of non-factual content but also increase user engagement. In addition, we analyze the trajectories of popular content in our simulations, and explore whether simulation agents' articulated reasoning for their social interactions truly aligns with their collective engagement patterns.

technical paper

## 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.

Audio often serves as an auxiliary modality in video understanding tasks of audio-visual large language models (LLMs), merely assisting in the comprehension of visual information. However, a thorough understanding of videos significantly depends on auditory information, as audio offers critical context, emotional cues, and semantic meaning that visual data alone often lacks. This paper proposes an audio-centric video understanding benchmark AVUT to evaluate the video comprehension capabilities of multimodal LLMs with a particular focus on auditory information. AVUT introduces a suite of carefully designed audio-centric tasks, holistically testing the understanding of both audio content and audio-visual interactions in videos. Moreover, this work points out the text shortcut problem that largely exists in other benchmarks where the correct answer can be found from question text alone without needing videos. AVUT addresses this problem by proposing a answer permutation-based filtering mechanism. A thorough evaluation across a diverse range of open-source and proprietary multimodal LLMs is performed, followed by the analyses of deficiencies in audio-visual LLMs. Demos are available at https://anonymous.4open.science/r/ACVUBench.

Audio-centric Video Understanding Benchmark without Text Shortcut

Unmanned Aerial Vehicle (UAV) Vision-and-Language Navigation (VLN) is vital for applications such as disaster response, logistics delivery, and urban inspection. However, existing methods often struggle with insufficient multimodal fusion, weak generalization, and poor interpretability. To address these challenges, we propose FlightGPT, a novel UAV VLN framework built upon Vision-Language Models (VLMs) with powerful multimodal perception capabilities. We design a two-stage training pipeline: first, Supervised Fine-Tuning (SFT) using high-quality demonstrations to improve initialization and structured reasoning; then, Group Relative Policy Optimization (GRPO) algorithm, guided by a composite reward that considers goal accuracy, reasoning quality, and format compliance, to enhance generalization and adaptability. Furthermore, FlightGPT introduces a Chain-of-Thought (CoT)-based reasoning mechanism to improve decision interpretability. Extensive experiments on the city-scale dataset CityNav demonstrate that FlightGPT achieves state-of-the-art performance across all scenarios, with a 9.22\% higher success rate than the strongest baseline in unseen environments. Our implementation is publicly available.

FlightGPT: Towards Generalizable and Interpretable UAV Vision-and-Language Navigation with Vision-Language Models

Text-to-image models frequently fail to achieve perfect alignment with textual prompts, particularly in maintaining proper semantic binding between semantic elements in the given prompt. Existing approaches typically require costly retraining or focus on only correctly generating the attributes of entities (entity-attribute binding), ignoring the cruciality of correctly generating the relations between entities (entity-relation-entity binding), resulting in unsatisfactory semantic binding performance. In this work, we propose a novel training-free method R-Bind that simultaneously improves both entity-attribute and entity-relation-entity binding. Our method introduces three inference-time optimization losses that adjust attention maps during generation. Comprehensive evaluations across multiple datasets demonstrate our approach's effectiveness, validity, and flexibility in enhancing semantic binding without additional training.

R-Bind: Unified Enhancement of Attribute and Relation Binding in Text-to-Image Diffusion Models

Large Language Models (LLMs) can improve commonsense reasoning through generating intermediate knowledge. However, the effectiveness of this knowledge introspection is not always guaranteed. This paper first systematically investigates and reveals an **introspection paradox**: while simple introspection tends to benefit weaker models, it often degrades the performance of stronger ones, particularly on simpler tasks. Our deep analysis indicates that this paradox arises from a complex interplay among model capability, task difficulty and the quality of generated knowledge. Further interpretability analysis reveals the origins of low-quality knowledge generation. To better employ introspected knowledge in LLM, this paper proposes a training-free **Adaptive Introspection Strategy** that operates in two stages using only the model's internal states: **Knowledge Detection**, which dynamically identifies and discards potentially low-quality knowledge, and **Knowledge Regeneration**, which employs attention smoothing to guide the model away from harmful failure modes during knowledge generation. Extensive experiments on five Llama models with different sizes and eight commonsense reasoning benchmarks demonstrate that our approach effectively mitigates the limitations of standard introspection and has consistent performance gains across almost all settings.

Why and How LLMs Benefit from Knowledge Introspection in Commonsense Reasoning

Dataset search is a specialized information retrieval task. In the emerging scenario of Dataset Search with Examples (DSE), the user submits a query and a few target datasets that are known to be relevant as examples. The retrieved datasets are expected to be relevant to the query and also similar to the target datasets. Distinguished from existing text-based retrievers, we propose a graph-based approach GraDaSE. Besides the textual metadata of the datasets, we identify their provenance-based and topic-based relationships to construct a graph, and jointly encode their structural and textual information for ranking candidate datasets. GraDaSE outperforms a variety of strong baselines on two test collections, including DataFinder-E that we construct.

GraDaSE: Graph-Based Dataset Search with Examples

We propose Confidence-guided Refinement Reasoning (C2R), a novel training-free framework applicable to question-answering (QA) tasks across text, image, and video domains. C2R strategically constructs and refines sub-questions and their answers (sub-QAs), deriving a better confidence score for the target answer. C2R first curates a subset of sub-QAs to explore diverse reasoning paths, then compares the confidence scores of the resulting answer candidates to select the most reliable final answer. Since C2R relies solely on confidence scores derived from the model itself, it can be seamlessly integrated with various existing QA models, demonstrating consistent performance improvements across diverse models and benchmarks. Furthermore, we provide essential yet underexplored insights into how leveraging sub-QAs affects model behavior, specifically analyzing the impact of both the quantity and quality of sub-QAs on achieving robust and reliable reasoning.

Confidence-guided Refinement Reasoning for Zero-shot Question Answering

Redundancy of visual tokens in multi-modal large language models (MLLMs) significantly reduces their computational efficiency. Recent approaches, such as resamplers and summarizers, have sought to reduce the number of visual tokens, but at the cost of visual reasoning ability. To address this, we propose LEO-Mini, a novel MLLM that significantly reduces the number of visual tokens and simultaneously boosts visual reasoning capabilities. For efficiency, LEO-Mini incorporates CoTR, a novel token reduction module to consolidate a large number of visual tokens into a smaller set of tokens, using the similarity between visual tokens, text tokens, and a compact learnable query. For effectiveness, to scale up the model's ability with minimal computational overhead, LEO-Mini employs MMoE, a novel mixture of multi-modal experts module. MMoE employs a set of LoRA experts with a novel router to switch between them based on the input text and visual tokens instead of only using the input hidden state. MMoE also includes a general LoRA expert that is always activated to learn general knowledge for LLM reasoning. For extracting richer visual features, MMoE employs a set of vision experts trained on diverse domain-specific data. To demonstrate LEO-Mini's improved efficiency and performance, we evaluate it against existing efficient MLLMs on various benchmark vision-language tasks.

LEO-MINI: An Efficient Multimodal Large Language Model using Conditional Token Reduction and Mixture of Multi-Modal Experts

Recent advances, such as DeepSeek R1-Zero, highlight the effectiveness of incentive training, a reinforcement learning paradigm that computes rewards solely based on the final answer part of a language model’s output, thereby encouraging the generation of intermediate reasoning steps. However, these methods fundamentally rely on external verifiers, which limits their applicability to domains like mathematics and coding, where such verifiers are readily available. Although reward models can serve as verifiers, they require high-quality annotated data and are costly to train. In this work, we propose textbfNOVER, textbftextitNO-VERifier Reinforcement Learning, a general reinforcement learning framework that requires only standard supervised fine-tuning data with no need for an external verifier. NOVER enables incentive training across a wide range of text-to-text tasks and outperforms the model of the same size distilled from large reasoning models such as DeepSeek R1 671B by 7.7\%. Moreover, the flexibility of NOVER enables new possibilities for optimizing large language models, such as inverse incentive training.

NOVER: Incentive Training for Language Models via Verifier-Free Reinforcement Learning

Grounded Multimodal Named Entity Recognition (GMNER) is a new information extraction task. It requires models to extract named entities and ground them to real-world visual objects. Previous methods, relying on domain-specific fine-tuning, struggle with unseen multimodal entities due to limited knowledge and generalization. Recently, multimodal large language models (MLLMs) have demonstrated strong open-set abilities. However, their performance is hindered by the lack of in-domain knowledge due to costly training for GMNER datasets. To address these limitations, we propose **UnCo**, a two-stage Uncertainty-driven Collaborative framework that leverages the complementary strengths of small fine-tuned models and MLLMs. Specifically, **in stage one**, we equip the small model with a unified uncertainty estimation (UE) for multimodal entities. This enables the small model to express textit"I do not know" when recognizing unseen entities beyond its capabilities. To obtain more robust UE performance, we introduce a modality representation alignment module that mitigates inter-modality bias. Predictions with high uncertainty are then filtered and delegated to the MLLM. **In stage two**, an Uncertainty-aware Hierarchical Correction mechanism guides the MLLM to refine uncertain predictions using its open-domain knowledge. Ultimately, UnCo effectively retains the in-domain knowledge of small models while utilizing the capabilities of MLLMs to handle unseen samples. Extensive experiments demonstrate UnCo's effectiveness on two GMNER benchmarks.

UnCo: Uncertainty-Driven Collaborative Framework of Large and Small Models for Grounded Multimodal NER

Large Language Models (LLMs) often hallucinate in question answering (QA) tasks. A key yet underexplored factor contributing to this is the temporality of questions -- whether they are evergreen (answers remain stable over time) or mutable (answers change). In this work, we introduce EverGreenQA, the first multilingual QA dataset with evergreen labels, supporting both evaluation and training. Using EverGreenQA, we benchmark 12 modern LLMs to assess whether they encode question temporality explicitly (via verbalized judgments) or implicitly (via uncertainty signals). We also train EG-E5, a lightweight multilingual classifier that achieves SoTA performance on this task. Finally, we demonstrate the practical utility of evergreen classification across three applications: improving self-knowledge estimation, filtering QA datasets, and explaining GPT-4o’s retrieval behavior.

Downloads

Next from EMNLP 2025

Audio-centric Video Understanding Benchmark without Text Shortcut

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES