Beyond the Surface: Measuring Self-Preference in LLM Judgments
Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学首都人工智能学院）；Huawei Poisson Lab（华为Poisson实验室）；Beijing Key Laboratory of Research on Large Models and Intelligent Governance（北京大型模型与智能治理重点实验室）
Zhi-Yuan Chen, Hao Wang, Xinyu Zhang, Enrui Hu, Yankai Lin
Recent studies show that large language models (LLMs) exhibit self-preference bias when serving as judges, meaning they tend to favor their own responses over those generated by other models. Existing methods typically measure this bias by calculating the difference between the scores a judge model assigns to its own responses and those it assigns to responses from other models. However, this approach conflates self-preference bias with response quality, as higher-quality responses from the judge model may also lead to positive score differences, even in the absence of bias. To address this issue, we introduce gold judgments as proxies for the actual quality of responses and propose the DBG score, which measures self-preference bias as the difference between the scores assigned by the judge model to its own responses and the corresponding gold judgments. Since gold judgments reflect true response quality, the DBG score mitigates the confounding effect of response quality on bias measurement. Using the DBG score, we conduct comprehensive experiments to assess self-preference bias across LLMs of varying versions, sizes, and reasoning abilities. Additionally, we investigate two factors that influence and help alleviate self-preference bias: response text style and the post-training data of judge models. Finally, we explore potential underlying mechanisms of self-preference bias from an attention-based perspective. Our code and data are available at https://github.com/zhiyuanc2001/self-preference.
Mitigating Judgment Preference Bias in Large Language Models through Group-Based Polling
School of Computer Science and Engineering, Northeastern University, Shenyang, China（东北大学计算机科学与工程学院，中国沈阳）；Department of Computer Science and Technology, Tsinghua University, Beijing, China（清华大学计算机科学与技术系，中国北京）；Software College, Northeastern University, Shenyang, China（东北大学软件学院，中国沈阳）；Huawei, China（华为，中国）
State Key Laboratory for Multimedia Information Processing, School of Computer Science, Peking University（信息处理国家重点实验室，计算机学院，北京大学）；StepFun；Xi’an Jiaotong University（西安交通大学）
From Blind Guess to Informed Judgment: Teaching LLMs to Evaluate Materials by Building Knowledge-Augmented Preference Signals
School of Computer Engineering & Science, Shanghai University（上海大学计算机工程与科学学院）；Center of Materials Informatics and Data Science, Materials Genome Institute, Shanghai University（上海大学材料信息与数据科学中心）；Key Laboratory of Silicate Cultural Relics Conservation (Shanghai University), Ministry of Education, China（教育部硅酸盐文化 relics 保护重点实验室（上海大学））；Shanghai Institute for Advanced Communication and Data Science, Shanghai University（上海大学高级通信与数据科学研究院）
Measuring Judgment Quality in Natural-Language Explanations: Evidence from Forecasting Tournaments
Forecasting Research Institute（预测研究所）；Good Judgment Inc（Good Judgment公司）；University of Toronto（多伦多大学）；Vector Institute for Artificial Intelligence（向量人工智能研究所）；Stanford University（斯坦福大学）；School of Arts and Sciences & Wharton, University of Pennsylvania（宾夕法尼亚大学文理学院与沃顿商学院）；Federal Reserve Bank of Chicago（芝加哥联邦储备银行）；Federal Reserve System（联邦储备系统）
JPO: Juris Policy Optimization for Structured Legal Reasoning in Criminal Judgment Prediction
Tencent（腾讯）；Peking University（北京大学）；Tsinghua University（清华大学）；City University of Hong Kong（香港城市大学）；University of California（加利福尼亚大学）；University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；Zhejiang University（浙江大学）；University of Hong Kong（香港大学）
Large-scale moral machine experiment on large language models
Department of Bioscience and Bioinformatics, Kyushu Institute of Technology（九州工科大学生物科学与生物信息学系）；Faculty of Engineering and Technology, Multimedia University（多媒体大学工程与技术学院）；Data Science and AI Research Center, Kyushu Institute of Technology（九州工科大学数据科学与人工智能研究中心）
Rubrics as an Attack Surface: Stealthy Preference Drift in LLM Judges
University of North Carolina at Chapel Hill（北卡罗来纳大学教堂山分校）；Carnegie Mellon University（卡内基梅隆大学）；Yale University（耶鲁大学）；The University of Texas at Austin（德克萨斯大学奥斯汀分校）
Austrian Research Institute for Artificial Intelligence (OFAI)（奥地利人工智能研究所）；Technische Universität Darmstadt（达姆施塔特工业大学）
Most humour processing systems to date make at best discrete, coarse-grained distinctions between the comical and the conventional, yet such notions are better conceptualized as a broad spectrum. In this paper, we present a probabilistic approach, a variant of Gaussian process preference learning (GPPL), that learns to rank and rate the humorousness of short texts by exploiting human preference judgments and automatically sourced linguistic annotations. We apply our system, which is similar to one that had previously shown good performance on English-language one-liners annotated with pairwise…
