CMU researchers are presenting 50 papers on the Thirtieth Convention on Empirical Strategies in Pure Language Processing (EMNLP 2025), held from November 4 \u2013 9 in Suzhou, China. This contains 27 papers in the principle convention, 19 papers within the Findings observe, 2 system demonstrations papers, and a couple of business observe papers. This weblog submit offers aggregated details about EMNLP 2025 papers printed by CMU researchers.<\/p>\n
Key areas addressed are visualized under (representing 30 of the 50 whole papers), illustrating the breadth of NLP and machine studying analysis being performed at CMU : <\/p>\n Observe<\/i>: All info on this submit has been obtained by the ACL Anthology API <\/a> and the EMNLP 2025 Presentation Data spreadsheet. Please contact CMU ML Weblog editors if you need any info added or modified.<\/p>\n Principal Convention Papers<\/a><\/strong><\/p>\n Findings Papers<\/a><\/strong><\/p>\n System Demonstrations<\/a><\/strong><\/p>\n Trade Observe Papers<\/a><\/strong><\/p>\n MolErr2Fix: Benchmarking LLM Trustworthiness in Chemistry through Modular Error Detection, Localization, Clarification, and Correction<\/a><\/strong> Spontaneous Giving and Calculated Greed in Language Fashions<\/a><\/strong> Artificial Socratic Debates: Analyzing Persona Results on Ethical Choice and Persuasion Dynamics<\/a><\/strong> Social Genome: Grounded Social Reasoning Skills of Multimodal Fashions<\/a><\/strong> VisualWebInstruct: Scaling up Multimodal Instruction Knowledge by Net Search<\/a><\/strong> Figuring out & Interactively Refining Ambiguous Person Targets for Knowledge Visualization Code Technology<\/a><\/strong> Persona-Augmented Benchmarking: Evaluating LLMs Throughout Various Writing Types<\/a><\/strong> Estimating LLM Consistency: A Person Baseline vs Surrogate Metrics<\/a><\/strong> Humanizing Machines: Rethinking LLM Anthropomorphism By way of a Multi-Degree Framework of Design<\/a><\/strong> Calibrating LLMs for Textual content-to-SQL Parsing by Leveraging Sub-clause Frequencies<\/a><\/strong> Not-Simply-Scaling Legal guidelines: In the direction of a Higher Understanding of the Downstream Influence of Language Mannequin Design Selections<\/a><\/strong> Rewarding the Unlikely: Lifting GRPO Past Distribution Sharpening<\/a><\/strong> Agentic-R1: Distilled Twin-Technique Reasoning<\/a><\/strong> SOCIAL SCAFFOLDS: A Generalization Framework for Social Understanding Duties<\/a><\/strong> Looking for the Most Human-like Emergent Language<\/a><\/strong> PhoniTale: Phonologically Grounded Mnemonic Technology for Typologically Distant Language Pairs<\/a><\/strong> Anecdoctoring: Automated Crimson-Teaming Throughout Language and Place<\/a><\/strong> CIE: Controlling Language Mannequin Textual content Generations Utilizing Steady Indicators<\/a><\/strong> Desk-R1: Inference-Time Scaling for Desk Reasoning Duties<\/a><\/strong> Grounding Multilingual Multimodal LLMs With Cultural Data<\/a><\/strong> Phrases Like Knives: Backstory-Customized Modeling and Detection of Violent Communication<\/a><\/strong> On the Tremendous-Grained Planning Skills of VLM Net Brokers<\/a><\/strong> An Empirical Examine on Robust-Weak Mannequin Collaboration for Repo-level Code Technology<\/a><\/strong> Summarizing Speech: A Complete Survey<\/a><\/strong> MoR: Higher Dealing with Various Queries with a Combination of Sparse, Dense, and Human Retrievers<\/a><\/strong> Morpheme Induction for Emergent Language<\/a><\/strong> Language Fashions Could be Effectively Steered through Minimal Embedding Layer Transformations<\/a><\/strong> FicSim: A Dataset for Multi-Faceted Semantic Similarity in Lengthy-Type Fiction<\/a><\/strong> ResearchArena: Benchmarking Giant Language Fashions\u2019 Means to Accumulate and Manage Data as Analysis Brokers<\/a><\/strong> SimBA: Simplifying Benchmark Evaluation Utilizing Efficiency Matrices Alone<\/a><\/strong> mrCAD: Multimodal Communication to Refine Pc-aided Designs<\/a><\/strong> Let Them Down Straightforward! Contextual Results of LLM Guardrails on Person Perceptions and Preferences<\/a><\/strong> Linear Steerability in Language Fashions: When It Emerges and How It Evolves<\/a><\/strong> Predicting Language Fashions\u2019 Success at Zero-Shot Probabilistic Prediction<\/a><\/strong> BenchMAX: A Complete Multilingual Analysis Suite for Giant Language Fashions<\/a><\/strong> FLAIRR-TS \u2013 Forecasting LLM-Brokers with Iterative Refinement and Retrieval for Time Collection<\/a><\/strong> Giant Language Mannequin Brokers in Finance: A Survey Bridging Analysis, Follow, and Actual-World Deployment<\/a><\/strong> VisCoder: Tremendous-Tuning LLMs for Executable Python Visualization Code Technology<\/a><\/strong> cAST: Enhancing Code Retrieval-Augmented Technology with Structural Chunking through Summary Syntax Tree<\/a><\/strong> GAMIC: Graph-Aligned Molecular In-context Studying for Molecule Evaluation through LLMs<\/a><\/strong> SVeritas: Benchmark for Sturdy Speaker Verification underneath Various Circumstances<\/a><\/strong> CAARMA: Class Augmentation with Adversarial Mixup Regularization<\/a><\/strong> Bridging the Enhancing Hole in LLMs: FineEdit for Exact and Focused Textual content Modifications<\/a><\/strong> Mitigate One, Skew One other? Tackling Intersectional Biases in Textual content-to-Picture Fashions<\/a><\/strong> Aligning Dialogue Brokers with World Suggestions through Giant Language Mannequin Multimodal Reward Decomposition<\/a><\/strong> TreeBoN: Enhancing Inference-Time Alignment with Speculative Tree-Search and Finest-of-N Sampling<\/a><\/strong> AgentDiagnose: An Open Toolkit for Diagnosing LLM Agent Trajectories<\/a><\/strong> BioGraphia: A LLM-Assisted Organic Pathway Graph Annotation Platform<\/a><\/strong>![\"\"](\"https:\/\/blog.ml.cmu.edu\/wp-content\/uploads\/2025\/11\/cmu_emnlp_2025_final_50papers-1024x716.png\")
<\/figure>\nDesk of Contents<\/h2>\n
Principal Convention Papers<\/h2>\n
Particular Theme: Interdisciplinary Recontextualization of NLP<\/h3>\n
Yuyang Wu, Jinhui Ye, Shuhao Zhang, Lu Dai, Yonatan Bisk, Olexandr Isayev<\/span><\/p>\n
Yuxuan Li, Hirokazu Shirado<\/span><\/p>\n
Jiarui Liu, Yueqi Track, Yunze Xiao, Mingqian Zheng, Lindia Tjuatja, Jana Schaich Borg, Mona T. Diab, Maarten Sap<\/span><\/p>\nMultimodality and Language Grounding to Imaginative and prescient, Robotics and Past<\/h3>\n
Leena Mathur, Marian Qian, Paul Pu Liang, Louis-Philippe Morency<\/span><\/p>\n
Yiming Jia, Jiachen Li, Xiang Yue, Bo Li, Ping Nie, Kai Zou, Wenhu Chen<\/span><\/p>\n
Mert Inan, Anthony Sicilia, Alex Xie, Saujas Vaduguru, Daniel Fried, Malihe Alikhani<\/span><\/p>\nSources and Analysis<\/h3>\n
Kimberly Truong, Riccardo Fogliato, Hoda Heidari, Steven Wu<\/span><\/p>\nHuman-AI Interplay\/Cooperation<\/h3>\n
Xiaoyuan Wu, Weiran Lin, Omer Akgul, Lujo Bauer<\/span><\/p>\n
Yunze Xiao, Lynnette Hui Xian Ng, Jiarui Liu, Mona T. Diab<\/span><\/p>\nInterpretability, Mannequin Enhancing, Transparency, and Explainability<\/h3>\n
Terrance Liu, Shuyi Wang, Daniel Preotiuc-Pietro, Yash Chandarana, Chirag Gupta<\/span><\/p>\n
Emmy Liu, Amanda Bertsch, Lintang Sutawika, Lindia Tjuatja, Patrick Fernandes, Lara Marinov, Michael Chen, Shreya Singhal, Carolin Lawrence, Aditi Raghunathan, Kiril Gashteovski, Graham Neubig<\/span><\/p>\nMathematical, Symbolic, and Logical Reasoning in NLP<\/h3>\n
Andre Wang He, Daniel Fried, Sean Welleck<\/span><\/p>\n
Weihua Du, Pranjal Aggarwal, Sean Welleck, Yiming Yang<\/span><\/p>\nGeneralizability and Switch<\/h3>\n
Ritam Dutt, Carolyn Rose, Maarten Sap<\/span><\/p>\n
Brendon Boldt, David R. Mortensen<\/span><\/p>\nNLP Functions<\/h3>\n
Sana Kang, Myeongseok Gwon, Su Younger Kwon, Jaewook Lee, Andrew Lan, Bhiksha Raj, Rita Singh<\/span><\/p>\nSecurity and Alignment in LLMs<\/h3>\n
Alejandro Cuevas, Saloni Sprint, Bharat Kumar Nayak, Dan Vann, Madeleine I. G. Daepp<\/span><\/p>\nPure Language Technology<\/h3>\n
Vinay Samuel, Harshita Diddee, Yiming Zhang, Daphne Ippolito<\/span><\/p>\nQuery Answering<\/h3>\n
Zheyuan Yang, Lyuhao Chen, Arman Cohan, Yilun Zhao<\/span><\/p>\nMultilinguality and Language Variety<\/h3>\n
Jean De Dieu Nyandwi, Yueqi Track, Simran Khanuja, Graham Neubig<\/span><\/p>\nComputational Social Science, Cultural Analytics, and NLP for Social Good<\/h3>\n
Jocelyn J Shen, Akhila Yerukola, Xuhui Zhou, Cynthia Breazeal, Maarten Sap, Hae Gained Park<\/span><\/p>\nAI\/LLM Brokers<\/h3>\n
Surgan Jandial, Yinong Oliver Wang, Andrea Bajcsy, Fernando De la Torre<\/span><\/p>\nCode Fashions<\/h3>\n
Shubham Gandhi, Atharva Naik, Yiqing Xie, Carolyn Rose<\/span><\/p>\nSummarization<\/h3>\n
Fabian Retkowski, Maike Z\u00fcfle, Andreas Sudmann, Dinah Pfau, Shinji Watanabe, Jan Niehues, Alexander Waibel<\/span><\/p>\nRetrieval-Augmented Language Fashions<\/h3>\n
Jushaan Singh Kalra, Xinran Zhao, To Eun Kim, Fengyu Cai, Fernando Diaz, Tongshuang Wu<\/span><\/p>\nPhonology, Morphology and Phrase Segmentation<\/h3>\n
Brendon Boldt, David R. Mortensen<\/span><\/p>\nLow-resource Strategies for NLP<\/h3>\n
Diogo Tavares, David Semedo, Alexander Rudnicky, Joao Magalhaes<\/span><\/p>\nFindings Papers<\/h2>\n
Particular Theme: Interdisciplinary Recontextualization of NLP<\/h3>\n
Natasha Johnson, Amanda Bertsch, Maria-Emil Deal, Emma Strubell<\/span><\/p>\nSources and Analysis<\/h3>\n
Hao Kang, Chenyan Xiong<\/span><\/p>\n
Nishant Subramani, Alfredo Gomez, Mona T. Diab<\/span><\/p>\n
William P McCarthy, Saujas Vaduguru, Karl D.d. Willis, Justin Matejka, Judith E Fan, Daniel Fried, Yewen Pu<\/span><\/p>\nHuman-AI Interplay\/Cooperation<\/h3>\n
Mingqian Zheng, Wenjia Hu, Patrick Zhao, Motahhare Eslami, Jena D. Hwang, Faeze Brahman, Carolyn Rose, Maarten Sap<\/span><\/p>\nInterpretability, Mannequin Enhancing, Transparency, and Explainability<\/h3>\n
Jianshu She, Xinyue Li, Eric P. Xing, Zhengzhong Liu, Qirong Ho<\/span><\/p>\n
Kevin Ren, Santiago Cortes-Gomez, Carlos Miguel Pati\u00f1o, Ananya Joshi, Ruiqi Lyu, Jingjing Tang, Alistair Turcan, Khurram Yamin, Steven Wu, Bryan Wilder<\/span><\/p>\nMultilinguality and Language Variety<\/h3>\n
Xu Huang, Wenhao Zhu, Hanxu Hu, Conghui He, Lei Li, Shujian Huang, Fei Yuan<\/span><\/p>\nAI\/LLM Brokers<\/h3>\n
Gunjan Jalori, Preetika Verma, Sercan O Arik<\/span><\/p>\n
Yifei Dong, Fengyi Wu, Kunlin Zhang, Yilong Dai, Sanjian Zhang, Wanghao Ye, Sihan Chen, Zhi-Qi Cheng<\/span><\/p>\nCode Fashions<\/h3>\n
Yuansheng Ni, Ping Nie, Kai Zou, Xiang Yue, Wenhu Chen<\/span><\/p>\nRetrieval-Augmented Language Fashions<\/h3>\n
Yilin Zhang, Xinran Zhao, Zora Zhiruo Wang, Chenyang Yang, Jiayi Wei, Tongshuang Wu<\/span><\/p>\n
Ali Al Lawati, Jason S Lucas, Zhiwei Zhang, Prasenjit Mitra, Suhang Wang<\/span><\/p>\nSpeech Processing and Spoken Language Understanding<\/h3>\n
Massa Baali, Sarthak Bisht, Francisco Teixeira, Kateryna Shapovalenko, Rita Singh, Bhiksha Raj<\/span><\/p>\n
Massa Baali, Xiang Li, Hao Chen, Syed Abdul Hannan, Rita Singh, Bhiksha Raj<\/span><\/p>\nSemantics: Lexical, Sentence-Degree Semantics, Textual Inference, and Different Areas<\/h3>\n
Yiming Zeng, Wanhao Yu, Zexin Li, Tao Ren, Yu Ma, Jinghan Cao, Xiyan Chen, Tingting Yu<\/span><\/p>\nEthics, Bias, and Equity<\/h3>\n
Pushkar Shukla, Aditya Chinchure, Emily Diana, Alexander Tolbert, Kartik Hosanagar, Vineeth N. Balasubramanian, Leonid Sigal, Matthew A. Turk<\/span><\/p>\nDialogue and Interactive Techniques<\/h3>\n
Dong Gained Lee, Hae Gained Park, Cynthia Breazeal, Louis-Philippe Morency<\/span><\/p>\nLLM Effectivity<\/h3>\n
Jiahao Qiu, Yifu Lu, Yifan Zeng, Jiacheng Guo, Jiayi Geng, Chenhao Zhu, Xinzhe Juan, Ling Yang, Huazheng Wang, Kaixuan Huang, Yue Wu, Mengdi Wang<\/span><\/p>\nSystem Demonstrations<\/h2>\n
Tianyue Ou, Wanyao Guo, Apurva Gandhi, Graham Neubig, Xiang Yue<\/span><\/p>\n
Xi Xu, Sumin Jo, Adam Officer, Angela Chen, Yufei Huang, Lei Li<\/span><\/p>\nTrade Observe Papers<\/h2>\n