Austria

Measuring the generalization ability of Large Language Models (LLMs) is challenging due to data contamination. As models grow and computation becomes cheaper, ensuring tasks and test cases are unseen during training phases will become nearly impossible. We argue that knowledge-retrieval and reasoning tasks are not ideal for measuring generalization, as LLMs are not trained for specific tasks. Instead, we propose user behavior prediction, also a key aspect of personalization, as a theoretically sound, scalable, and robust alternative. We introduce a novel framework for this approach and test it on movie and music recommendation datasets for GPT-4o, GPT-4o-mini, and Llama-3.1-8B-Instruct. Results align with our framework’s predictions, showing GPT-4o outperforms GPT-4o-mini and Llama, though all models have much room for improvement, especially Llama.

ACL 2025

User Behavior Prediction as a Generic, Robust, Scalable, and Low-Cost Evaluation Strategy for Estimating Generalization in LLMs

task hierarchy

recommendation

personalization

generalization

## 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." 

*Message from the General Chair, Dirk Hovy*

EMNLP 2025

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

poster

### Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Message from the General Chair: 
*It is my great pleasure and honor to welcome you to the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025), held in beautiful Vienna, Austria, from July 27 to August 1, 2025. ACL2025continues our field’s tradition of excellence in scholarship, innovation, and inclusivity, and I am deeply grateful to the many volunteers who have worked tirelessly to bring this event to life.* 
[Read more](https://drive.google.com/file/d/1GI_hvOpjswAuYdUTromfeDiPpCcqidwg/view?usp=sharing)

To access this event page, you need to log in with the **email address you registered with**. Access credentials will be sent to your email from Underline - subject line "Welcome to ACL 2025". Please be sure to check your spam email folder if you do not see an email confirmation right away.

Please log in

To access this event page, you need to log in with the **email address you registered with**. 

Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Large language models (LLMs) excel in high-resource languages but struggle with low-resource languages (LRLs), particularly those spoken by minority communities in China, such as Tibetan, Uyghur, Kazakh, and Mongolian. To systematically track the progress in these languages, we introduce MiLiC-Eval, a benchmark designed for minority languages in China, featuring 24K instances across 9 tasks. MiLiC-Eval focuses on underrepresented writing systems and provides a fine-grained assessment of linguistic and problem-solving skills. Our evaluation reveals that LLMs perform poorly on syntax-intensive tasks and multi-script languages. We further demonstrate how MiLiC-Eval can help advance LRL research in handling diverse writing systems and understanding the process of language adaptation.

MiLiC-Eval: Benchmarking Multilingual LLMs for China's Minority Languages

Training large language models (LLMs) to follow instructions has significantly enhanced their ability to tackle unseen tasks. However, despite their strong generalization capabilities, instruction-following LLMs encounter difficulties when dealing with tasks that require domain knowledge. This work introduces a specialized instruction fine-tuning for the domain of computational argumentation (CA). The goal is to enable an LLM to effectively tackle any unseen CA tasks while preserving its generalization capabilities. Reviewing existing CA research, we crafted natural language instructions for 105 CA tasks to this end. On this basis, we developed a CA-specific benchmark for LLMs that allows for a comprehensive evaluation of LLMs' capabilities in solving various CA tasks. We synthesized 52k CA-related instructions, adapting the self-instruct process to train a CA-specialized instruction-following LLM. Our experiments suggest that CA-specialized instruction fine-tuning significantly enhances the LLM on both seen and unseen CA tasks. At the same time, performance on the general NLP tasks of the SuperNI benchmark remains stable.

ArgInstruct: Specialized Instruction Fine-Tuning for Computational Argumentation

Large Language Models (LLMs) have shown impressive moral reasoning abilities. Yet they often diverge when confronted with complex, multi-factor moral dilemmas. To address these discrepancies, we propose a framework that synthesizes multiple LLMs' moral judgments into a collectively formulated moral judgment, realigning models that deviate significantly from this consensus. Our aggregation mechanism fuses continuous moral acceptability scores (beyond binary labels) into a collective probability, weighting contributions by model reliability. For misaligned models, a targeted embedding-optimization procedure fine-tunes token embeddings for moral concepts, minimizing JS divergence to the consensus while preserving semantic integrity. Experiments on a large-scale social moral dilemma dataset show our approach builds robust consensus and improves individual model fidelity. These findings highlight the value of data-driven moral alignment across multiple models and its potential for safer, more consistent AI systems.

Probabilistic Aggregation and Targeted Embedding Optimization for Collective Moral Reasoning in Large Language Models

Large language models (LLMs) have exhibited remarkable versatility and adaptability, while their widespread adoption across various applications also raises critical safety concerns. This paper focuses on the impact of backdoored LLMs. Traditional backdoor injection methods are primarily limited to yes-or-no discriminative tasks, leading users to underestimate the potential risks of backdoored LLMs. Given the inherently generative nature of LLMs, this paper reveals that a generative backdoor injected into LLMs can expose the true safety risks in their application. We propose an editing-based generative backdoor, named MEGen, aiming to expand the backdoor to generative tasks in a unified format of any text-to any text, leading to natural generations with a specific intention. Experiments show that MEGen achieves a high attack success rate by adjusting only a small set of local parameters with few-shot samples. Notably, we show that the backdoored model, when triggered, can freely output pre-set dangerous information while completing downstream tasks. Our work highlights that MEGen enables backdoors in LLMs to exhibit generative capabilities, causing potential safety risks by altering the generative style.

MEGen: Generative Backdoor into Large Language Models via Model Editing

Complex multi-hop question answering requires large language models (LLMs) not only to retrieve external knowledge but also to reason over the retrieved information in order to arrive at the final solution. This involves two key challenges: (i) how to effectively explore the solution space and generate more potentially correct solution candidates, and (ii) how to select the optimal solution from multiple solution candidates, both of which require a training-free approach without introducing a more powerful teacher model. To address these challenges, we propose Retrieval-Augmented Monte Carlo Tree Self-Play with Reasoning Consistency (RASPberry), which introduces a more flexible action-level sampling granularity compared to existing methods, leverages Monte Carlo Tree Search for efficient solution space exploration, and utilizes an enhanced version of reasoning consistency to guide the selection of the optimal solution. Experimental results demonstrate that our proposed RASPberry effectively tackles the two challenges outlined above, achieving more efficient RAG inference-time scaling. Our code is available at https://github.com/BaixuanLi/RASPberry.

RASPberry: Retrieval-Augmented Monte Carlo Tree Self-Play with Reasoning Consistency for Multi-Hop Question Answering

Correctly identifying characters and substrings of words should be a basic but essential ability of any Language Model that aims to proficiently understand and produce language. Despite so, the majority of Pre-trained Language Models (PLMs) are "character-blind" and struggle in spelling tasks, although they still seem to acquire some character knowledge during pre-training, a phenomenon dubbed Spelling Miracle. To shed light on this phenomenon, we systematically evaluate a range of PLMs with different parameter sizes using a controlled binary substring identification task. Through a series of experiments, we propose the first comprehensive investigation on where, when, and how a PLMs develop awareness of characters and substrings, with a particular linguistic focus on morphemic units such as prefixes, suffixes, and roots.

Beyond the Spelling Miracle: Investigating Substring Awareness in Character-Blind Language Models

Emotion Cause Triplet Extraction in Multimodal Conversations (MECTEC) has recently gained significant attention in social media analysis, aiming to extract emotion utterances, cause utterances, and emotion categories simultaneously. However, the scarcity of related datasets, with only one published dataset featuring highly uniform dialogue scenarios, hinders model development in this field. To address this, we introduce MECAD, the first multimodal, multi-scenario MECTEC dataset, comprising 989 conversations from 56 TV series spanning a wide range of dialogue contexts. In addition, existing MECTEC methods fail to explicitly model emotional and causal contexts and neglect the fusion of semantic information at different levels, leading to performance degradation. In this paper, we propose M3HG, a novel model that explicitly captures emotional and causal contexts and effectively fuses contextual information at both inter- and intra-utterance levels via a multimodal heterogeneous graph. Extensive experiments demonstrate the effectiveness of M3HG compared with existing state-of-the-art methods. Codes are available at https://anonymous.4open.science/r/M3HG-6B34.

M3HG: Multimodal, Multi-scale, and Multi-type Node Heterogeneous Graph for Emotion Cause Triplet Extraction in Conversations

Predicting video popularity is often framed as a supervised learning task, relying heavily on meta-information and aggregated engagement data. However, video popularity is shaped by complex cultural and social factors that such approaches often overlook. We argue that Large Language Models (LLMs), with their deep contextual awareness, can better capture these nuances. To bridge the gap between pixel-based video data and token-based LLMs, we convert frame-level visuals into sequential text representations using Vision-Language Models. This enables LLMs to process multimodal content—titles, frame-based descriptions, and captions—capturing both engagement intensity (view count) and geographic spread (number of countries where a video trends). On 13,639 popular videos, a supervised neural network using content embeddings achieves 80\% accuracy, while our LLM-based approach reaches 82\% without fine-tuning. Combining the neural network's predictions with the LLM further improves accuracy to 85.5\%. Moreover, the LLM generates interpretable, attribute-based explanations for its predictions. Manual validations confirm the quality of these hypotheses and address concerns about hallucinations in the video-to-text conversion process. Overall, our findings suggest that LLMs, equipped with text-based multimodal representations, offer a powerful, interpretable, and data-efficient solution for tasks requiring rich contextual insight, such as video popularity prediction.

Large Language Models Are Natural Video Popularity Predictors

Meeting summarization suffers from limited high-quality data, mainly due to privacy restrictions and expensive collection processes. We address this gap with FAME, a dataset of 500 meetings in English and 300 in German produced by MIMIC, our new multi-agent meeting synthesis framework that generates meeting transcripts on a given knowledge source by defining psychologically grounded participant profiles, outlining the conversation, and orchestrating a large language model (LLM) debate. A modular post-processing step refines these outputs, mitigating potential repetitiveness and overly formal tones, ensuring coherent, credible dialogues at scale. We also propose a psychologically grounded evaluation framework assessing naturalness, social behavior authenticity, and transcript difficulties. Human assessments show that FAME approximates real-meeting spontaneity (4.5/5 in naturalness), preserves speaker-centric challenges (3/5 in spoken language), and introduces richer information-oriented difficulty (4/5 points in difficulty). These findings show FAME is a good and scalable proxy for real-world meeting conditions. It enables new test scenarios for meeting summarization research and other conversation-centric applications in tasks requiring conversation data or simulating social scenarios under behavioral constraints.

You need to MIMIC to get FAME: Solving Meeting Transcript Scarcity with a Multi-Agent Conversations

The theoretical code-switching (CS) literature provides numerous pointwise investigations that aim to explain patterns in CS, i.e. why bilinguals switch language in certain positions in a sentence more often than in others. A resulting consensus is that CS acceptability can be explained by the syntax of the contributing languages. There is however no large-scale, multi-language, cross-phenomena experiment that tests this claim. When designing such an experiment, we need to make sure that the system that is predicting where bilinguals tend to switch has access only to syntactic information. We provide such an experiment here. Results show that syntax alone is sufficient for an automatic system to distinguish between sentences in minimal pairs of CS, to the same degree as bilingual humans. Furthermore, the learnt syntactic patterns generalise well to unseen language pairs.

Downloads

Next from ACL 2025

MiLiC-Eval: Benchmarking Multilingual LLMs for China's Minority Languages

Similar lecture

Are the Reasoning Models Good at Automated Essay Scoring?

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES