Macro: Enhancing Multilingual Counterfactual Explanations through Alignment-as-Preference Optimization
Technische Universität Berlin（柏林技术大学）；German Research Center for Artificial Intelligence (DFKI)（德国人工智能研究中心）；University of Duisburg-Essen（杜伊斯堡- Essen大学）；LMU Munich（慕尼黑大学）；Munich Center for Machine Learning (MCML)（慕尼黑机器学习中心）；Saarland Informatics Campus（萨尔兰州信息学校区）；BIFOLD – Berlin Institute for the Foundations of Learning and Data（柏林学习与数据基础研究院）；Centre for European Research in Trusted AI (CERTAIN)（可信人工智能欧洲研究中心）
Self-generated counterfactual explanations (SCEs) are minimally modified inputs (minimality) generated by large language models (LLMs) that flip their own predictions (validity), offering a causally grounded approach to unraveling black-box LLM behavior. Yet extending them beyond English remains challenging: existing methods struggle to produce valid SCEs in non-dominant languages, and a persistent trade-off between validity and minimality undermines explanation quality. We introduce Macro, a preference alignment framework that applies Direct Preference Optimization (DPO) to multilingual SCE generation, using a composite scoring function to construct preference pairs that effectively translate the trade-off into measurable preference signals. Experiments across four LLMs and seven typologically diverse languages show that Macro improves validity by 12.55\% on average over the chain-of-thought baseline without degrading minimality, while avoiding the severe minimality violations of the translation-based baseline. Compared to supervised fine-tuning, Macro achieves superior performance on both metrics, confirming that explicit preference optimization is essential for balancing this trade-off. Further analyses reveal that Macro increases cross-lingual perturbation alignment and mitigates common generation errors. Our results highlight preference optimization as a promising direction for enhancing multilingual model explanations.
Parallel Universes, Parallel Languages: A Comprehensive Study on LLM-based Multilingual Counterfactual Example Generation
Technische Universität Berlin（柏林工业大学）；German Research Center for Artificial Intelligence (DFKI)（德国人工智能研究中心（DFKI））；University of Marburg（马尔堡大学）；LMU Munich（慕尼黑大学）；Munich Center for Machine Learning (MCML)（慕尼黑机器学习中心（MCML））；BIFOLD – Berlin Institute for the Foundations of Learning and Data（柏林学习与数据基础研究所（BIFOLD））
Reliable Multilingual Orthopedic Decision Support from Clinical Narratives: Language-Aware Adaptation and Verification-Guided Deferral
School of Computer Science, Wuhan University（武汉大学计算机学院）；School of Software, Nanjing University of Information Science and Technology (NUIST)（南京信息工程大学软件学院）；Department of Orthopedic, Bahawal Victoria Hospital（巴哈瓦尔维多利亚医院骨科部门）
Counterfactual statements describe events that did not or cannot take place. We consider the problem of counterfactual detection (CFD) in product reviews. For this purpose, we annotate a multilingual CFD dataset from Amazon product reviews covering counterfactual statements written in English, German, and Japanese languages. The dataset is unique as it contains counterfactuals in multiple languages, covers a new application area of e-commerce reviews, and provides high quality professional annotations. We train CFD models using different text representation methods and classifiers. We find tha…
MCEval: A Dynamic Framework for Fair Multilingual Cultural Evaluation of LLMs
Zhejiang University（浙江大学）；School of Engineering, Westlake University（西溪大学工程学院）；Southern University of Science and Technology（南方科技大学）；Institute of Advanced Technology, Westlake Institute for Advanced Study（西溪研究院先进技术研究所）
National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy Sciences（中国科学院自动化研究所模式识别国家重点实验室）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；College of Electrical and Information Engineering, Hunan University（湖南大学电气与信息工程学院）
Scientific documents often contain a large number of acronyms. Disambiguation of these acronyms will help researchers better understand the meaning of vocabulary in the documents. In the past, thanks to large amounts of data from English literature, acronym task was mainly applied in English literature. However, for other low-resource languages, this task is difficult to obtain good performance and receives less attention due to the lack of large amount of annotation data. To address the above issue, this paper proposes an new method for acronym disambiguation, named as ADBCMM, which can signi…
CAMO: An Agentic Framework for Automated Causal Discovery from Micro Behaviors to Macro Emergence in LLM Agent Simulations
College of Intelligence and Computing, Tianjin University（天津大学智能计算学院）；Tianjin Key Laboratory of Healthy Habitat and Smart Technology（天津健康人居环境与智能技术重点实验室）；Laboratory of Computation and Analytics of Complex Management Systems, Tianjin University（复杂管理系统计算与分析实验室）
MOSAIC: A Multilingual, Taxonomy-Agnostic, and Computationally Efficient Approach for Radiological Report Classification
Department of Computer Science, University of Copenhagen（哥本哈根大学计算机科学系）；Neurobiology Research Unit, Copenhagen University Hospital（哥本哈根大学医院神经生物学研究单位）；Unumed Aps（Unumed公司）；Department of Diagnostic Radiology, Copenhagen University Hospital（哥本哈根大学医院诊断放射学系）；Department of Clinical Medicine, University of Copenhagen（哥本哈根大学临床医学系）；Cerebriu A/S（Cerebriu公司）；Institute for Human Genetics, Medical University of Innsbruck（因斯布鲁克医学大学人类遗传学研究所）
Mohamed bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学）；The Fin AI（金融人工智能公司）；FMI, Sofia University “St. Kliment Ohridski”（索非亚大学“圣克莱门特·奥赫里德斯基”金融数学与信息学系）；INSAIT, Sofia University “St. Kliment Ohridski”（索非亚大学“圣克莱门特·奥赫里德斯基”信息技术与自动化研究所）；University of Arizona（亚利桑那大学）；The University of Tokyo（东京大学）；Linköping University（林雪平大学）；McGill University（麦吉尔大学）；Mila, Quebec AI Institute（魁北克人工智能研究所）；Meiji Gakuin University（明治学院大学）
A Multimodal, Multilingual, and Multidimensional Pipeline for Fine-grained Crowdsourcing Earthquake Damage Evaluation
New York University（纽约大学）；University of South Florida（佛罗里达州立大学）；Google LLC（谷歌公司）；University of California, Santa Barbara（加州大学圣巴巴拉分校）；Massachusetts Institute of Technology（麻省理工学院）
MetaPlate: Counterfactual-Guided RAG-LLM Tool for Personalized Food Recommendation and Hyperglycemia Prevention
College of Health Solutions, Arizona State University（亚利桑那州立大学健康解决方案学院）；School of Computing and Augmented Intelligence, Arizona State University（亚利桑那州立大学计算与增强智能学院）
University of Oregon（俄勒冈大学）；Princeton University（普林斯顿大学）；Adobe Research（Adobe研究院）
Acronym extraction is the task of identifying acronyms and their expanded forms in texts that is necessary for various NLP applications. Despite major progress for this task in recent years, one limitation of existing AE research is that they are limited to the English language and certain domains (i.e., scientific and biomedical). As such, challenges of AE in other languages and domains is mainly unexplored. Lacking annotated datasets in multiple languages and domains has been a major issue to hinder research in this area. To address this limitation, we propose a new dataset for multilingual …
University of Mannheim（曼海姆大学）；Bocconi University（博科尼大学）
Unfair stereotypical biases (e.g., gender, racial, or religious biases) encoded in modern pretrained language models (PLMs) have negative ethical implications for widespread adoption of state-of-the-art language technology. To remedy for this, a wide range of debiasing techniques have recently been introduced to remove such stereotypical biases from PLMs. Existing debiasing methods, however, directly modify all of the PLMs parameters, which -- besides being computationally expensive -- comes with the inherent risk of (catastrophic) forgetting of useful language knowledge acquired in pretrainin…
Birla Institute of Technology and Science, Pilani（皮拉尼贝拉理工与科学学院）
Online social networks are ubiquitous and user-friendly. Nevertheless, it is vital to detect and moderate offensive content to maintain decency and empathy. However, mining social media texts is a complex task since users don't adhere to any fixed patterns. Comments can be written in any combination of languages and many of them may be low-resource. In this paper, we present our system for the LT-EDI shared task on detecting homophobia and transphobia in social media comments. We experiment with a number of monolingual and multilingual transformer based models such as mBERT along with a data a…
In this paper, we describe our approach to utilize pre-trained BERT models with Convolutional Neural Networks for sub-task A of the Multilingual Offensive Language Identification shared task (OffensEval 2020), which is a part of the SemEval 2020. We show that combining CNN with BERT is better than using BERT on its own, and we emphasize the importance of utilizing pre-trained language models for downstream tasks. Our system, ranked 4th with macro averaged F1-Score of 0.897 in Arabic, 4th with score of 0.843 in Greek, and 3rd with score of 0.814 in Turkish. Additionally, we present ArabicBERT, …
Johns Hopkins University（约翰斯·霍普金斯大学）；New York University（纽约大学）
Structural probing work has found evidence for latent syntactic information in pre-trained language models. However, much of this analysis has focused on monolingual models, and analyses of multilingual models have employed correlational methods that are confounded by the choice of probing tasks. In this study, we causally probe multilingual language models (XGLM and multilingual BERT) as well as monolingual BERT-based models across various languages; we do this by performing counterfactual perturbations on neuron activations and observing the effect on models' subject-verb agreement probabili…
BenSyc: Benchmarking Conversational Sycophancy and Human Alignment in LLMs for Bengali Contexts
University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；Virginia Tech（弗吉尼亚理工大学）；Bangladesh University of Engineering and Technology（孟加拉工程与技术大学）；BRAC University（BRAC大学）；University of Toronto（多伦多大学）
University of Wolverhampton（伍尔弗汉普顿大学）；Rochester Institute of Technology（罗切斯特理工学院）
Offensive content is pervasive in social media and a reason for concern to companies and government organizations. Several studies have been recently published investigating methods to detect the various forms of such content (e.g. hate speech, cyberbullying, and cyberaggression). The clear majority of these studies deal with English partially because most annotated datasets available contain English data. In this paper, we take advantage of available English datasets by applying cross-lingual contextual word embeddings and transfer learning to make predictions in low-resource languages. We pr…
LLMs for automatic annotation of Mandarin narrative transcripts
Shanghai Jiao Tong University（上海交通大学）；University of California San Diego（加州大学圣地亚哥分校）；The Hong Kong Polytechnic University（香港理工大学）；National Research Center for Language and Well-Being（语言与幸福感国家研究中心）；City University of Hong Kong（香港城市大学）
We present MultiCoNER, a large multilingual dataset for Named Entity Recognition that covers 3 domains (Wiki sentences, questions, and search queries) across 11 languages, as well as multilingual and code-mixing subsets. This dataset is designed to represent contemporary challenges in NER, including low-context scenarios (short and uncased text), syntactically complex entities like movie titles, and long-tail entity distributions. The 26M token dataset is compiled from public resources using techniques such as heuristic-based sentence sampling, template extraction and slotting, and machine tra…
University of Wolverhampton（伍尔弗汉普顿大学）；Rochester Institute of Technology（罗切斯特理工学院）
Offensive content is pervasive in social media and a reason for concern to companies and government organizations. Several studies have been recently published investigating methods to detect the various forms of such content (e.g. hate speech, cyberbulling, and cyberaggression). The clear majority of these studies deal with English partially because most annotated datasets available contain English data. In this paper, we take advantage of English data available by applying cross-lingual contextual word embeddings and transfer learning to make predictions in languages with less resources. We …
Indian Institute of Technology, Kharagpur（印度理工学院卡拉格普尔分校）；National Institute of Technology Raipur（赖布尔国家理工学院）
The number of increased social media users has led to a lot of people misusing these platforms to spread offensive content and use hate speech. Manual tracking the vast amount of posts is impractical so it is necessary to devise automated methods to identify them quickly. Large language models are trained on a lot of data and they also make use of contextual embeddings. We fine-tune the large language models to help in our task. The data is also quite unbalanced; so we used a modified cross-entropy loss to tackle the issue. We observed that using a model which is fine-tuned in hindi corpora pe…
QU-NLP at CheckThat! 2025: Multilingual Subjectivity in News Articles Detection using Feature-Augmented Transformer Models with Sequential Cross-Lingual Fine-Tuning
University of Washington（华盛顿大学）；The University of Wisconsin-Madison（威斯康星大学麦迪逊分校）；The University of Texas at Austin（德克萨斯大学奥斯汀分校）
While the NLP community is generally aware of resource disparities among languages, we lack research that quantifies the extent and types of such disparity. Prior surveys estimating the availability of resources based on the number of datasets can be misleading as dataset quality varies: many datasets are automatically induced or translated from English data. To provide a more comprehensive picture of language resources, we examine the characteristics of 156 publicly available NLP datasets. We manually annotate how they are created, including input text and label sources and tools used to buil…
