Austria

Large Language Models (LLMs) demonstrate strong abilities in common-sense reasoning and interactive decision-making, but often struggle with complex, long-horizon planning tasks. Recent techniques have sought to structure LLM outputs using control flow and code to improve planning performance. However, code-based approaches can be error-prone and insufficient for handling ambiguous or unstructured data. To address these challenges, we propose REPL-Plan, an LLM planning approach that is fully code-expressive (it can utilize all the benefits of code) while also being dynamic (it can flexibly adapt from errors and use the LLM for soft reasoning). In REPL-Plan, an LLM solves tasks by interacting with a Read-Eval-Print Loop (REPL), which iteratively executes and evaluates code, similar to language shells or interactive code notebooks, allowing the model to flexibly correct errors and handle tasks dynamically. We demonstrate that REPL-Plan achieves strong results across various planning domains compared to previous methods.

ACL 2025

Interactive and Expressive Code-Augmented Planning with Large Language Models

llm agents

code

agents

large language model

decision making

planning

reinforcement learning

### Welcome!
The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) will take place in Bangkok, Thailand from August 11th to 16th, 2024. Our Virtual Poster Sessions will take place online Thursday, August 22, 2024.

You are required to register for this event. **Please register [here](https://2024.aclweb.org/registration). **

If you have already registered, please check your inbox for an email from Underline granting you access to ACL 2024 content.

Please register!

ACL 2024

The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) will take place in Bangkok, Thailand from August 11th to 16th, 2024. More information will be announced soon.

technical paper

### 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)

Human evaluation is crucial for assessing rapidly evolving language models but is influenced by annotator proficiency and task design. This study explores the integration of comparative judgment into human annotation for machine translation (MT) and evaluates three annotation setups—point-wise Multidimensional Quality Metrics (MQM), side-by-side (S×S) MQM, and its simplified version S×S relative ranking (RR). In MQM, annotators mark error spans with categories and severity levels. S×S MQM extends MQM to pairwise error annotation for two translations of the same input, while S×S RR focuses on selecting the better output without labeling errors.

Key findings are: (1) the S×S settings achieve higher inter-annotator agreement than MQM; (2) S×S MQM enhances inter-translation error marking consistency compared to MQM by, on average, 38.5% for explicitly compared MT systems and 19.5% for others; (3) all annotation settings return stable system rankings, with S×S RR offering a more efficient alternative to (S×S) MQM; (4) the S×S settings highlight subtle errors overlooked in MQM without altering absolute system evaluations.

To spur further research, we will release the triply annotated datasets comprising 377 ZhEn and 104 EnDe annotation examples, each covering 10 systems.

Enhancing Human Evaluation in Machine Translation with Comparative Judgement

As artificial intelligence surpasses human capabilities in text generation, the necessity to authenticate the origins of AI-generated content has become paramount. Unbiased watermarks offer a powerful solution by embedding statistical signals into language model-generated text without distorting the quality. In this paper, we introduce MCmark, a family of unbiased, Multi-Channel-based watermarks. MCmark works by partitioning the model's vocabulary into segments and promoting token probabilities within a selected segment based on a watermark key. We demonstrate that MCmark not only preserves the original distribution of the language model but also offers significant improvements in detectability and robustness over existing unbiased watermarks. Our experiments with widely-used language models demonstrate an improvement in detectability of over 10% using MCmark, compared to existing state-of-the-art unbiased watermarks. This advancement underscores MCmark's potential in enhancing the practical application of watermarking in AI-generated texts.

Improved Unbiased Watermark for Large Language Models

Recently, multimodal large language models (MLLMs) have been extensively explored in video question answering. However, most existing assessments focus on natural videos, overlooking synthetic videos (e.g., AI-generated content). Meanwhile, some works in video generation rely on MLLMs to evaluate the quality of generated videos, but the capabilities of MLLMs on AIGC videos remain largely underexplored. To address this, we propose a new benchmark, VQ-Eval, which introduces four tasks—coherence validation, error awareness, error type detection, and reasoning evaluation—to comprehensively evaluate the abilities of MLLMs on AIGC videos. We evaluate 13 frontier MLLMs on VQ-Eval and find that even the best-performing model, GPT-4.1, struggles to achieve consistently good performance across all tasks. This highlights the challenging nature of our benchmark. Additionally, to investigate the practical applications of VQ-Eval in improving video generation, we design a re-prompt pipeline, demonstrating that aligning MLLMs more closely with human feedback can benefit the video generation.

VQ-Eval: Evaluating Multimodal LLMs for Generating Feedback on AIGC Videos

Multilingual models often treat language diversity as a problem of data imbalance, overlooking structural variation. We introduce the *Morphological Index* (MoI), a typologically grounded metric that quantifies how strongly a language relies on surface morphology for noun classification. Building on MoI, we propose *MoI-MoE*, a Mixture of Experts model that routes inputs based on morphological structure. Evaluated on 10 Bantu languages—a large, morphologically rich and underrepresented family—MoI-MoE outperforms strong baselines, improving Swahili accuracy by 14 points on noun class recognition while maintaining performance on morphology-rich languages like Zulu. These findings highlight typological structure as a practical and interpretable signal for multilingual model adaptation.

Typology-Guided Adaptation for African NLP

To create culturally inclusive vision-language models (VLMs), developing a benchmark that tests their ability to address culturally relevant questions is essential. Existing approaches typically rely on human annotators, making the process labor-intensive and creating a cognitive burden in generating diverse questions. To address this, we propose a semi-automated framework for constructing cultural VLM benchmarks, specifically targeting multiple-choice QA. This framework combines human-VLM collaboration, where VLMs generate questions based on guidelines, a small set of annotated examples, and relevant knowledge, followed by a verification process by native speakers. We demonstrate the effectiveness of this framework through the creation of K-Viscuit, a dataset focused on Korean culture. Our experiments on this dataset reveal that open-source models lag behind proprietary ones in understanding Korean culture, highlighting key areas for improvement. We also present a series of further analyses, including human evaluation, augmenting VLMs with external knowledge, and the evaluation beyond multiple-choice QA. Our dataset is available at https://huggingface.co/datasets/ddehun/k-viscuit.

Evaluating Visual and Cultural Interpretation: The K-Viscuit Benchmark with Human-VLM Collaboration

This paper addresses the challenge of adapting language models to the jurisdiction-specific nature of legal corpora. Existing approaches—training separate models for each jurisdiction or using a single shared model—either fail to leverage common legal principles beneficial for low-resource settings or risk negative interference from conflicting jurisdictional interpretations. To overcome these limitations, we propose a parameter-efficient framework ProMALex, that first derives hierarchical relationships across jurisdictions and progressively inserts adapter modules across model layers based on jurisdictional similarity. This design allows modules in lower layers to be shared across jurisdictions, capturing common legal principles, while higher layers specialize through jurisdiction-specific adapters. Experimental results on two legal language modeling benchmarks demonstrate that ProMALex outperforms both fully shared and jurisdiction-specific models.

ProMALex: Progressive Modular Adapters for Multi-Jurisdictional Legal Language Modeling

Recent advancements in tree search algorithms guided by verifiers have significantly enhanced the reasoning capabilities of large language models (LLMs), but at the cost of increased computational resources. In this work, we identify two key challenges contributing to this inefficiency: $\textit{over-exploration}$ due to redundant states with semantically equivalent content, and $\textit{under-exploration}$ caused by high variance in verifier scoring leading to frequent trajectory switching. 
To address these issues, we propose FETCH -- an e${\bf f}$fici${\bf e}$nt ${\bf t}$ree sear${\bf ch}$ framework, which is a flexible, plug-and-play system compatible with various tree search algorithms.
Our framework mitigates over-exploration by merging semantically similar states using agglomerative clustering of text embeddings obtained from a fine-tuned SimCSE model. To tackle under-exploration, we enhance verifiers by incorporating temporal difference learning with adjusted $\lambda$-returns during training to reduce variance, and employing a verifier ensemble to aggregate scores during inference. Experiments on GSM8K, GSM-Plus, and MATH datasets demonstrate that our methods significantly improve reasoning accuracy and computational efficiency across four different tree search algorithms, paving the way for more practical applications of LLM-based reasoning. The code is available at https://github.com/DeepLearnXMU/Fetch.

Don't Get Lost in the Trees: Streamlining LLM Reasoning by Overcoming Tree Search Exploration Pitfalls

As Large Language Models (LLMs) increasingly participate in human-AI interactions, evaluating their Theory of Mind (ToM) capabilities - particularly their ability to track dynamic mental states - becomes crucial. While existing benchmarks assess basic ToM abilities, they predominantly focus on static snapshots of mental states, overlooking the temporal evolution that characterizes real-world social interactions. We present **DynToM**, a novel benchmark specifically designed to evaluate LLMs' ability to understand and track the temporal progression of mental states across interconnected scenarios. Through a systematic four-step framework, we generate 1,100 social contexts encompassing 5,500 scenarios and 78,100 questions, each validated for realism and quality. Our comprehensive evaluation of ten state-of-the-art LLMs reveals that their average performance underperforms humans by 44.7%, with performance degrading significantly when tracking and reasoning about the shift of mental states. This performance gap highlights fundamental limitations in current LLMs' ability to model the dynamic nature of human mental states.

Towards Dynamic Theory of Mind: Evaluating LLM Adaptation to Temporal Evolution of Human States

Large multimodal foundation models, particularly in the domains of language and vision, have significantly advanced various tasks, including robotics, autonomous driving, information retrieval, and grounding. However, many of these models perceive objects as indivisible, overlooking the components that constitute them. Understanding these components and their associated affordances provides valuable insights into an object's functionality, which is fundamental for performing a wide range of tasks. In this work, we introduce a novel real-world benchmark, InstructPart, comprising hand-labeled part segmentation annotations and task-oriented instructions to evaluate the performance of current models in understanding and executing part-level tasks within everyday contexts. Through our experiments, we demonstrate that task-oriented part segmentation remains a challenging problem, even for state-of-the-art Vision-Language Models (VLMs). In addition to our benchmark, we introduce a simple baseline that achieves a twofold performance improvement through fine-tuning with our dataset. With our dataset and benchmark, we aim to facilitate research on task-oriented part segmentation and enhance the applicability of VLMs across various domains, including robotics, virtual reality, information retrieval, and other related fields. Project website: https://zifuwan.github.io/InstructPart/.

InstructPart: Task-Oriented Part Segmentation with Instruction Reasoning

Generating step-by-step “chain-of-thought” rationales has proven effective for improving the performance of large language models on complex reasoning tasks. However, applying such techniques to structured tasks, such as text-to-SQL, remains largely unexplored. In this paper, we introduce Self-Taught Reasoner for text-to-SQL (STaR-SQL), a novel approach that reframes SQL query generation as a reasoning-driven process. Our method prompts the LLM to produce detailed reasoning steps for SQL queries and fine-tunes it on rationales that lead to correct outcomes. Unlike traditional methods, STaR-SQL dedicates additional test-time computation to reasoning, thereby positioning LLMs as spontaneous reasoners rather than mere prompt-based agents. To further scale the inference process, we incorporate an outcome-supervised reward model (ORM) as a verifier, which enhances SQL query accuracy. Experimental results on the challenging Spider benchmark demonstrate that STaR-SQL significantly improves text-to-SQL performance, achieving an execution accuracy of 86.6\%. This surpasses a few-shot baseline by 31.6\% and a baseline fine-tuned to predict answers directly by 18.0\%. Additionally, STaR-SQL outperforms agent-like prompting methods that leverage more powerful yet closed-source models such as GPT-4. These findings underscore the potential of reasoning-augmented training for structured tasks and open the door to extending self-improving reasoning models to text-to-SQL generation and beyond.

Downloads

Next from ACL 2025

Enhancing Human Evaluation in Machine Translation with Comparative Judgement

Similar lecture

Open Grounded Planning: Challenges and Benchmark Construction