Austria

Online discourse is increasingly trapped in a vicious cycle where polarizing language fuels
toxicity and vice versa. Identity, one of the most divisive issues in modern politics, often
increases polarization. Yet, prior NLP research has mostly treated toxicity and polarization as
separate problems. In Indonesia, the world’s third-largest democracy, this dynamic threatens democratic discourse, particularly in online spaces. We argue that polarization and toxicity must be studied in relation to each other. To this end, we present a novel multi-label Indonesian dataset annotated for toxicity, polarization, and annotator demographic information. Benchmarking with BERT-base models and large language models (LLMs) reveals that polarization cues improve toxicity classification and vice versa. Including demographic context further enhances polarization classification performance.

ACL 2025

A Multi-Labeled Dataset for Indonesian Discourse: Examining Toxicity, Polarization, and Demographics Information

indonesian nlp

toxicity

polarization

identity

demographic information

### 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)

poster

Large language models (LLMs) are increasingly impacting human society, particularly in textual information. Based on more than 30,000 papers and 1,000 presentations from machine learning conferences, we examined and compared the words used in writing and speaking, representing the first large-scale study of how LLMs influence the two main modes of verbal communication and expression within the same group of people. Our empirical results show that LLM-style words such as significant have been used more frequently in abstracts and oral presentations. The implicit impact on human expression like writing and speaking is beginning to emerge and is likely to grow in the future. We take the first step in building an automated monitoring platform to record its longitudinal changes to call attention to the implicit influence and ripple effect of LLMs on human society.

The Impact of Large Language Models in Academia: from Writing to Speaking

Inductive reasoning, a cornerstone of human cognition, enables generalization from limited data but hasn't yet been fully achieved by large language models (LLMs). While modern LLMs excel at reasoning tasks, their ability to maintain stable and consistent rule abstraction under imperfect observations remains underexplored. To fill this gap, in this work, we introduce Robust Rule Induction, a task that evaluates LLMs' capability in inferring rules from data that are fused with noisy examples. To address this task, we further propose Sample-steered Rule Refinement (SRR), a method enhancing reasoning stability via observation diversification and execution-guided feedback. Experiments across arithmetic, cryptography, and list functions reveal: (1) SRR outperforms other methods with minimal performance degradation under noise; (2) Despite slight accuracy variation, LLMs exhibit instability under noise (e.g., 0% accuracy change with only 70% consistent score); (3) Counterfactual task gaps highlight LLMs' reliance on memorized patterns over genuine abstraction. Our findings challenge LLMs' reasoning robustness, revealing susceptibility to hypothesis drift and pattern overfitting, while providing empirical evidence critical for developing human-like inductive systems. Code and data are available at https://github.com/HKUST-KnowComp/Robust-Rule-Induction.

Patterns Over Principles: The Fragility of Inductive Reasoning in LLMs under Noisy Observations

With the vast expansion of content on social media platforms, analyzing and comprehending online discourse has become increasingly complex. This paper introduces LLMTaxo, a novel framework leveraging large language models for the automated construction of taxonomy of factual claims from social media by generating topics from multi-level granularities. This approach aids stakeholders in more effectively navigating the social media landscapes. We implement this framework with different models across three distinct datasets and introduce specially designed taxonomy evaluation metrics for a comprehensive assessment. With the evaluations from both human evaluators and GPT-4, the results indicate that LLMTaxo effectively categorizes factual claims from social media, and reveals that certain models perform better on specific datasets.

LLMTaxo: Leveraging Large Language Models for Constructing Taxonomy of Factual Claims from Social Media

Large Vision Language Models (LVLMs) have shown impressive performance on various vision-language tasks. However, while objects in natural scenes inevitably exhibit visual variations in position, scale, orientation, and context due to changes in viewpoint and environment, the robustness of LVLMs to these fundamental visual variations remains largely unexplored. To address this gap, we introduce V²R-Bench, a comprehensive benchmark framework for evaluating Visual Variation Robustness of LVLMs, which encompasses automated evaluation dataset generation and principled metrics for thorough robustness assessment. Through extensive evaluation of 13 LVLMs, we reveal a surprising vulnerability to visual variations, affecting even advanced models that excel at complex vision-language tasks yet significantly underperform on simple tasks like object recognition. Interestingly, these models exhibit a distinct visual position bias that contradicts theories of effective receptive fields and demonstrate a human-like visual acuity threshold. To identify the source of these vulnerabilities, we propose a systematic framework for component-level analysis, featuring a novel visualization approach for aligned visual features. Results show that these vulnerabilities stem from error accumulation in the pipeline architecture and inadequate multimodal alignment. Complementary experiments with synthetic data further demonstrate that these limitations are fundamentally architectural challenges, underscoring the need for architectural innovations in future LVLM designs.

V²R-Bench: Holistically Evaluating LVLM Robustness to Fundamental Visual Variations

Large Language Models (LLMs) face significant challenges when queried about long-tail knowledge, i.e., information that is rarely encountered during their training process. These difficulties arise due to the inherent sparsity of such data. Furthermore, LLMs often lack the ability to verify or ground their responses in authoritative sources, which can lead to plausible yet inaccurate outputs when addressing infrequent subject matter. Our work aims to investigate these phenomena by introducing KE-mHISTO, a multilingual benchmark for Entity Linking and Question Answering in the domain of historical music knowledge, available in both Italian and English. We demonstrate that KE-mHISTO provides significantly broader coverage of long-tail knowledge compared to existing alternatives. Moreover, it poses substantial challenges for state-of-the-art models. Our experiments reveal that smaller, multilingual models can achieve performance comparable to significantly larger counterparts, highlighting the potential of efficient, language-aware approaches for long-tail knowledge extraction. KE-mHISTO is available at: https://anonymous.4open.science/r/KE-MHISTO-6108.

KE-MHISTO: Towards a Multilingual Historical Knowledge Extraction Benchmark for Addressing the Long-Tail Problem

Electromyography (EMG) tables are crucial for diagnosing muscle and nerve disorders, and advancing the automation of EMG diagnostics is significant for improving medical efficiency. EMG tables contain extensive continuous numerical data, which current Large Language Models (LLMs) often struggle to interpret effectively. To address this issue, we propose EMGLLM, a data-to-text model specifically designed for medical examination tables. EMGLLM employs the EMG Alignment Encoder to simulate the process that doctors compare test values with reference values, aligning the data into word embeddings that reflect health degree. Additionally, we construct ETM, a dataset comprising 17,276 real cases and their corresponding diagnostic results, to support medical data-to-text tasks. Experimental results on ETM demonstrate that EMGLLM outperforms various baseline models in understanding EMG tables and generating high-quality diagnoses, which represents an effective paradigm for automatic diagnosis generation from medical examination table.

EMGLLM: Data-to-Text Alignment for Electromyogram Diagnosis Generation with Medical Numerical Data Encoding

Machine Unlearning (MU) is critical for removing private or hazardous information from deep learning models. While MU has advanced significantly in unimodal (text or vision) settings, multimodal unlearning (MMU) remains underexplored due to the lack of open benchmarks for evaluating cross-modal data removal. To address this gap, we introduce CLEAR, the first open-source benchmark designed specifically for MMU. CLEAR contains 200 fictitious individuals and 3,700 images linked with corresponding question-answer pairs, enabling a thorough evaluation across modalities. We conduct a comprehensive analysis of 11 MU methods (e.g., SCRUB, gradient ascent, DPO) across four evaluation sets, demonstrating that jointly unlearning both modalities outperforms single-modality approaches.

CLEAR: Character Unlearning in Textual and Visual Modalities

Although LLMs have shown great performance on Mathematics and Coding related reasoning tasks, the reasoning capabilities of LLMs regarding other forms of reasoning are still an open problem. Here, we examine the issue of reasoning from the perspective of claim verification. We propose a framework designed to break down any claim paired with evidence into atomic reasoning types that are necessary for verification. We use this framework to create RECV, the first claim verification benchmark, incorporating real-world claims, to assess the deductive and abductive reasoning capabilities of LLMs. The benchmark comprises of three datasets, covering reasoning problems of in creasing complexity. We evaluate three state of-the-art proprietary LLMs under multiple prompt settings. Our results show that while LLMs can address deductive reasoning prob lems, they consistently fail in cases of abductive reasoning. Moreover, we observe that enhancing LLMs with rationale generation is not always beneficial. Nonetheless, we find that generated rationales are semantically similar to those provided by humans, especially in deduc tive reasoning cases.

Assessing the Reasoning Capabilities of LLMs in the context of Evidence-based Claim Verification

Instruction following (IF) is a critical capability for large language models (LLMs). However, handling complex instructions with multiple constraints remains challenging. Previous methods typically select preference pairs based on the number of constraints they satisfy, introducing noise where chosen examples may fail to follow some constraints and rejected examples may excel in certain respects over the chosen ones. To address the challenge of aligning with multiple preferences, we propose a simple yet effective method called Reverse Preference Optimization (RPO). It mitigates noise in preference pairs by dynamically reversing the constraints within the instruction to ensure the chosen response is perfect, alleviating the burden of extensive sampling and filtering to collect perfect responses. Besides, reversal also enlarges the gap between chosen and rejected responses, thereby clarifying the optimization direction and making it more robust to noise. We evaluate RPO on two multi-turn IF benchmarks, Sysbench and Multi-IF, demonstrating average improvements over the DPO baseline of 4.6 and 2.5 points~(on Llama-3.1 8B), respectively. Moreover, RPO scales effectively across model sizes (8B to 70B parameters), with the 70B RPO model surpassing GPT-4o.

Reverse Preference Optimization for Complex Instruction Following

Large language models (LLMs) have been widely adopted due to their remarkable performance across various applications, driving the accelerated development of a large number of diverse models. However, these individual LLMs show limitations in generalization and performance on complex tasks due to inherent training biases, model size constraints, and the quality or diversity of pre-training datasets. A promising direction is to efficiently harness the diverse capabilities of LLMs to overcome these individual limitations. To address these limitations, we introduce a novel LLM selection algorithm called SelectLLM, which efficiently directs input queries to the most suitable subset of LLMs from a large pool, ensuring that the selected models collectively provide accurate responses. SelectLLM employs a multi-label classifier and policy based on the classifier's predictions and confidence scores in selecting an optimal, query-aware, and lightweight subset of LLMs. Our findings indicate that the proposed model outperforms existing ensemble-based baselines and achieves competitive performance with similarly sized top-performing LLMs while maintaining efficiency. Specifically, it achieves a huge reduction in inference latency on two challenging reasoning benchmarks: 13% on GSM8K and 70% on MMLU, compared to the top-performing baseline. Also, we establish a theoretical upper bound by an Oracle with LLMs and perform an in-depth linguistic analysis to understand the performance gap between the Oracle and SelectLLM.

Downloads

Next from ACL 2025

The Impact of Large Language Models in Academia: from Writing to Speaking

Similar lecture

(Non-Archival Submission) A Multi-Labeled Dataset for Indonesian Discourse: Examining Toxicity, Polarization, and Demographics Information