Can Knowledge Graphs Make Large Language Models More Trustworthy? An Empirical Study Over Open-ended Question Answering
机构 * National University of Singapore（新加坡国立大学） ； University of Cambridge（剑桥大学）
Comments This paper has been accepted by ACL 2025
期刊&会议
ACL
Annual Meeting of the Association for Computational Linguistics · 会议 · Natural Language Processing
LangSAMP: Language-Script Aware Multilingual Pretraining
机构 * Center for Information and Language Processing, LMU Munich（信息与语言处理中心，慕尼黑莱茵河大学） ； Munich Center for Machine Learning (MCML)（慕尼黑机器学习中心（MCML）） ； Bosch Center for Artificial Intelligence（博世人工智能中心）
Comments ACL 2025
MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark
Comments ACL 2025 Main
BlockPruner: Fine-grained Pruning for Large Language Models
机构 * School of Computer Science and Engineering, Sun Yat-sen University（中山大学计算机科学与工程学院） ； Meetyou AI Lab（Meetyou人工智能实验室）
Comments ACL 2025 Findings
FiDeLiS: Faithful Reasoning in Large Language Model for Knowledge Graph Question Answering
机构 * National University of Singapore（新加坡国立大学） ； University of Science and Technology of China（中国科学技术大学）
Comments This paper has been accepted by ACL 2025
Language Models are Universal Embedders
机构 * Harbin Institute of Technology（哈尔滨工业大学） ； The Hong Kong Polytechnic University（香港理工大学）
Comments XLLM Workshop, ACL 2025
LABO: Towards Learning Optimal Label Regularization via Bi-level Optimization
机构 * Department of Computer Science and Operations Research, Université de Montréal（蒙特利尔大学计算机科学与运筹学系） ； Huawei Noah’s Ark Lab, Canada（华为诺亚实验室，加拿大） ； Department of Statistics and Actuarial Science, University of Waterloo（滑铁卢大学统计与精算科学系）
Comments Accepted at ACL2023 (Findings)
Transfer of Structural Knowledge from Synthetic Languages
机构 * School of Computer Science & Engineering（计算机科学与工程学院） ； Constructor University（Constructor大学） ； CAIRO THWS ； Würzburg, Germany（德国魏尔堡）
Comments 10 pages, 3 figures and 3 tables to be published in ACL 2025 Workshop XLLM
Listen to the Context: Towards Faithful Large Language Models for Retrieval Augmented Generation on Climate Questions
机构 * Machine Learning and Human Language Technology, RWTH Aachen University, Germany（机器学习与人类语言技术，亚琛工业大学，德国） ； AppTek GmbH, Aachen, Germany（AppTek GmbH，亚琛，德国）
Comments Accepted at the ClimateNLP 2025 Workshop at ACL
GODBench: A Benchmark for Multimodal Large Language Models in Video Comment Art
Comments 69 pages, 66 figures, accepted by ACL 2025
Scaling Text-Rich Image Understanding via Code-Guided Synthetic Multimodal Data Generation
机构 * University of Pennsylvania（宾夕法尼亚大学） ； Allen Institute for Artificial Intelligence（人工智能研究所）
Comments Published in ACL 2025, project page: https://yueyang1996.github.io/cosyn/
SQL Injection Jailbreak: A Structural Disaster of Large Language Models
Comments Accepted by findings of ACL 2025
CoLA: Collaborative Low-Rank Adaptation
机构 * Zhejiang University（浙江大学）
Comments Accepted by ACL 2025, Findings
X-WebAgentBench: A Multilingual Interactive Web Benchmark for Evaluating Global Agentic System
Comments Accepted by ACL 2025 Findings
Leveraging Unit Language Guidance to Advance Speech Modeling in Textless Speech-to-Speech Translation
Comments Accepted to ACL 2025 Findings
Exploring In-Image Machine Translation with Real-World Background
机构 * School of Computer Science and Technology, Beijing Institute of Technology（北京理工大学计算机科学与技术学院） ； School of Computer Science and Engineering, Beihang University（北航计算机科学与工程学院）
Comments Accepted to ACL 2025 Findings. Code available at https://github.com/BITHLP/DebackX
Towards Explainable Temporal Reasoning in Large Language Models: A Structure-Aware Generative Framework
机构 * School of Computer Science, Wuhan University（武汉大学计算机学院） ； The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））
Comments In Findings of the Association for Computational Linguistics: ACL 2025
EcomScriptBench: A Multi-task Benchmark for E-commerce Script Planning via Step-wise Intention-Driven Product Association
机构 * Department of Computer Science and Engineering, HKUST, Hong Kong SAR, China（香港科技大学计算机科学与工程系） ； Amazon.com Inc（亚马逊公司） ； Department of Computer Science & Engineering, Texas A&M University（德克萨斯农工大学计算机科学与工程系）
Comments ACL2025
Can Large Language Models Understand Internet Buzzwords Through User-Generated Content
机构 * College of Computer Science, Sichuan University, China（四川大学计算机学院） ； Engineering Research Center of Machine Learning and Industry Intelligence, Ministry of Education, China（教育部机器学习与工业智能工程研究中心） ； JD.com, China（京东公司）
Comments ACL 2025 Main Paper. Our dataset and code are available at https://github.com/SCUNLP/Buzzword
In-Domain African Languages Translation Using LLMs and Multi-armed Bandits
机构 * Media Analysis Group, Sony Research India（索尼印度研究机构媒体分析组）
Journal ref AfricaNLP Workshop at ACL 2025
Instruction-Tuning Data Synthesis from Scratch via Web Reconstruction
机构 * The Hong Kong University of Science and Technology (Guangzhou)（香港科学与技术大学（广州）） ； The Hong Kong University of Science and Technology（香港科学与技术大学） ； Huawei Noah’s Ark Lab（华为诺亚实验室）
Comments 16 pages, 11 figures, 9 tables. ACL 2025 camera-ready version
GiFT: Gibbs Fine-Tuning for Code Generation
机构 * Nanyang Technological University（南洋理工大学）
Comments Accepted to ACL 2025
Analyzing the Effect of Linguistic Similarity on Cross-Lingual Transfer: Tasks and Experimental Setups Matter
机构 * Munich Center for Machine Learning（慕尼黑机器学习中心） ； Apple（苹果公司）
Comments ACL Findings 2025
MIRe: Enhancing Multimodal Queries Representation via Fusion-Free Modality Interaction for Multimodal Retrieval
机构 * Department of Artificial Intelligence, Korea University（人工智能系，韩国大学）
Comments Accepted to ACL 2025 (Findings)
Robust and Minimally Invasive Watermarking for EaaS
机构 * Tsinghua University（清华大学） ； The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））
Comments Accepted by ACL 2025
MARS: Benchmarking the Metaphysical Reasoning Abilities of Language Models with a Multi-task Evaluation Dataset
机构 * Department of Computer Science and Engineering, HKUST, Hong Kong SAR, China（计算机科学与工程系，香港科技大学，香港特别行政区，中国）
Comments ACL2025
KERL: Knowledge-Enhanced Personalized Recipe Recommendation using Large Language Models
机构 * Rensselaer Polytechnic Institute（拉特格斯理工学院）
Comments Accepted at ACL 2025
Neural Incompatibility: The Unbridgeable Gap of Cross-Scale Parametric Knowledge Transfer in Large Language Models
机构 * The Key Laboratory of Cognition and Decision Intelligence for Complex Systems, Institute of Automation, Chinese Academy of Sciences（认知与决策智能复杂系统重点实验室，自动化研究所，中国科学院） ； School of Artificial Intelligence, University of Chinese Academy of Sciences（人工智能学院，中国科学院大学）
Comments Accepted by ACL'25 Main. Code link: https://github.com/Trae1ounG/Neural_Incompatibility
WirelessMathBench: A Mathematical Modeling Benchmark for LLMs in Wireless Communications
机构 * School of Electrical and Electronic Engineering (EEE), Nanyang Technological University, Singapore（南洋理工大学电子与电气工程学院） ； Department of Computer and Information Engineering, Khalifa University, Abu Dhabi, UAE（哈利法大学计算机与信息工程学院）
Comments Accepted to ACL 2025 Findings
Cross-Lingual Optimization for Language Transfer in Large Language Models
机构 * Korea University（韩国大学）
Comments Accepted for publication at ACL 2025. Jungseob Lee and Seongtae Hong contributed equally to this work
