大模型对齐与安全

ENCORE: Entropy-guided Reward Composition for Multi-head Safety Reward Models

Xiaomin Li, Xupeng Chen, Jingxuan Fan, Eric Hanchen Jiang, Mingye Gao

专题命中 偏好对齐 ：safety(title,abstract)；alignment(abstract)；RLHF(abstract)；分类 cs.CL

Towards Reward Fairness in RLHF: From a Resource Allocation Perspective

Sheng Ouyang, Yulan Hu, Ge Chen, Qingyang Li, Fuzheng Zhang, Yong Liu

机构 * Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学人工智能学院） ； Beijing Key Laboratory of Research on Large Models and Intelligent Governance（北京大型模型与智能治理研究重点实验室） ； Engineering Research Center of Next-Generation Intelligent Search and Recommendation, MOE（下一代智能搜索与推荐工程技术研究中心，教育部） ； Kuaishou Technology（快手科技） ； University of Chinese Academy of Sciences（中国科学院大学） 纠错

专题命中 偏好对齐 ：RLHF(title,abstract)；alignment(abstract)；分类 cs.AI、cs.LG

Comments Accepted to ACL 2025

CONGRAD:Conflicting Gradient Filtering for Multilingual Preference Alignment

Jiangnan Li, Thuy-Trang Vu, Christian Herold, Amirhossein Tebbifakhr, Shahram Khadivi, Gholamreza Haffari

机构 * Department of Data Science and AI, Monash University, Australia（墨尔本大学数据科学与人工智能系） ； eBay Inc.（eBay公司） 纠错

专题命中 偏好对齐 ：alignment(title,abstract)；分类 cs.CL

The Polite Liar: Epistemic Pathology in Language Models

Bentley DeVilling

专题命中 偏好对齐 ：alignment(abstract)；RLHF(abstract)；分类 cs.CL、cs.AI、cs.CY

Comments 17 pages, 2 tables, Preprint - under review at AI & Society

CAPO: Confidence Aware Preference Optimization Learning for Multilingual Preferences

Rhitabrat Pokharel, Yufei Tao, Ameeta Agrawal

机构 * Department of Computer Science（计算机科学系） 纠错

专题命中 偏好对齐 ：alignment(abstract)；DPO(abstract)；分类 cs.CL、cs.AI

Comments Accepted at IJCNLP-AACL 2025 Findings

Interaction Dynamics as a Reward Signal for LLMs

Sian Gooding, Edward Grefenstette

机构 * Google DeepMind（谷歌DeepMind） 纠错

专题命中 偏好对齐 ：alignment(abstract)；分类 cs.CL、cs.AI、cs.LG

SparseRM: A Lightweight Preference Modeling with Sparse Autoencoder

Dengcan Liu, Jiahao Li, Zheren Fu, Yi Tu, Jiajun Li, Zhendong Mao, Yongdong Zhang

专题命中 偏好对齐 ：alignment(abstract)；分类 cs.CL、cs.AI

Comments 15pages,11figures,AAAI-26

Beyond Correctness: Confidence-Aware Reward Modeling for Enhancing Large Language Model Reasoning

Qianxi He, Qingyu Ren, Shanzhe Lei, Xuhong Wang, Yingchun Wang

机构 * Shanghai Artificial Intelligence Laboratory（上海人工智能实验室） ； Shanghai Key Laboratory of Data Science, School of Computer Science, Fudan University（上海数据科学关键实验室，计算机科学学院，复旦大学） 纠错

专题命中 偏好对齐 ：alignment(abstract)；分类 cs.AI、cs.LG

RESTL: Reinforcement Learning Guided by Multi-Aspect Rewards for Signal Temporal Logic Transformation

Yue Fang, Jin Zhi, Jie An, Hongshen Chen, Xiaohong Chen, Naijun Zhan

专题命中 偏好对齐 ：alignment(abstract)

Comments 12 pages, 4 figures

Alignment-Constrained Dynamic Pruning for LLMs: Identifying and Preserving Alignment-Critical Circuits

Dev Patel, Gabrielle Gervacio, Diekola Raimi, Kevin Zhu, Ryan Lagasse, Gabriel Grand, Ashwinee Panda, Maheep Chaudhary

专题命中 安全训练 ：alignment(title,abstract)；safety(abstract)；分类 cs.CL、cs.AI、cs.LG

STAR-1: Safer Alignment of Reasoning LLMs with 1K Data

Zijun Wang, Haoqin Tu, Yuhan Wang, Juncheng Wu, Yanqing Liu, Jieru Mei, Brian R. Bartoldson, Bhavya Kailkhura, Cihang Xie

机构 * UC Santa Cruz（加州大学圣克ruz分校） ； Google（谷歌） ； Lawrence Livermore National Laboratory（劳伦斯利弗莫尔国家实验室） 纠错

专题命中 安全训练 ：alignment(title,abstract)；safety(abstract)；分类 cs.CL、cs.AI

Provably Efficient Sample Complexity for Robust CMDP

Sourav Ganguly, Arnob Ghosh

机构 * New Jersey Institute of Technology（新泽西理工学院） 纠错

专题命中 安全训练 ：safety(abstract)；分类 cs.LG

KG-DF: A Black-box Defense Framework against Jailbreak Attacks Based on Knowledge Graphs

Shuyuan Liu, Jiawei Chen, Xiao Yang, Hang Su, Zhaoxia Yin

专题命中 越狱攻击 ：jailbreak(title,abstract)；分类 cs.AI

Why does weak-OOD help? A Further Step Towards Understanding Jailbreaking VLMs

Yuxuan Zhou, Yuzhao Peng, Yang Bai, Kuofeng Gao, Yihao Zhang, Yechao Zhang, Xun Chen, Tao Yu, Tao Dai, Shu-Tao Xia

专题命中 越狱攻击 ：alignment(abstract)；safety(abstract)；jailbreak(abstract)

RedCodeAgent: Automatic Red-teaming Agent against Diverse Code Agents

Chengquan Guo, Chulin Xie, Yu Yang, Zhaorun Chen, Zinan Lin, Xander Davies, Yarin Gal, Dawn Song, Bo Li

专题命中 越狱攻击 ：safety(abstract)；jailbreak(abstract)

Decoding Latent Attack Surfaces in LLMs: Prompt Injection via HTML in Web Summarization

Ishaan Verma, Arsheya Yadav

专题命中 提示注入 ：prompt injection(title,abstract)；分类 cs.AI

A Cautionary Tale About "Neutrally" Informative AI Tools Ahead of the 2025 Federal Elections in Germany

Ina Dormuth, Sven Franke, Marlies Hafer, Tim Katzke, Alexander Marx, Emmanuel Müller, Daniel Neider, Markus Pauly, Jérôme Rutinowski

专题命中 提示注入 ：alignment(abstract)；prompt injection(abstract)；分类 cs.AI、cs.LG

Journal ref Explainable Artificial Intelligence (xAI 2025). Communications in Computer and Information Science, vol. 2580, pp. 64-85. Springer, Cham (2026)

On the Role of Calibration in Benchmarking Algorithmic Fairness for Skin Cancer Detection

Brandon Dominique, Prudence Lam, Nicholas Kurtansky, Jochen Weber, Kivanc Kose, Veronica Rotemberg, Jennifer Dy

机构 * Northeastern University（东北大学） ； Memorial Sloan Kettering Cancer Center（纪念斯隆凯特琳癌症中心） 纠错

专题命中 幻觉与事实性 ：alignment(abstract)；分类 cs.LG

Comments 19 pages, 4 figures. Accepted for publication at the Journal of Machine Learning for Biomedical Imaging (MELBA) https://melba-journal.org/2025:027

Journal ref Machine.Learning.for.Biomedical.Imaging. 3 (2025)

Hard vs. Noise: Resolving Hard-Noisy Sample Confusion in Recommender Systems via Large Language Models

Tianrui Song, Wen-Shuo Chao, Hao Liu

专题命中 幻觉与事实性 ：alignment(abstract)；分类 cs.AI

Comments Accepted by AAAI 2026

SAFENLIDB: A Privacy-Preserving Safety Alignment Framework for LLM-based Natural Language Database Interfaces

Ruiheng Liu, XiaoBing Chen, Jinyu Zhang, Qiongwen Zhang, Yu Zhang, Bailong Yang

专题命中 隐私与版权 ：alignment(title,abstract)；safety(title)；DPO(abstract)；分类 cs.CL

Comments AAAI 2026 Extended Version

Integrating Artificial Intelligence into Operating Systems: A Survey on Techniques, Applications, and Future Directions

Yifan Zhang, Xinkui Zhao, Ziying Li, Guanjie Cheng, Jianwei Yin, Lufei Zhang, Zuoning Chen

机构 * School of Software Technology, Zhejiang University（浙江大学软件学院） ； College of Computer Science and Technology, Zhejiang University（浙江大学计算机科学与技术学院） ； State Key Laboratory of Mathematical Engineering and Advanced Computing（数学工程与先进计算国家重点实验室） ； Chinese Academy of Engineering（中国工程院） ； College of Cyber Security, Jinan University（暨南大学网络安全学院） 纠错

专题命中 隐私与版权 ：safety(abstract)；分类 cs.AI

Comments 68 pages,9 figures

A Decentralized Retrieval Augmented Generation System with Source Reliabilities Secured on Blockchain

Yining Lu, Wenyi Tang, Max Johnson, Taeho Jung, Meng Jiang

专题命中 隐私与版权 ：trustworthy(abstract)；分类 cs.CL

How Brittle is Agent Safety? Rethinking Agent Risk under Intent Concealment and Task Complexity

Zihan Ma, Dongsheng Zhu, Shudong Liu, Taolin Zhang, Junnan Liu, Qingqiu Li, Minnan Luo, Songyang Zhang, Kai Chen

机构 * School of Computer Science and Technology Xi’an Jiaotong University China（西安交通大学计算机科学与技术学院） ； Shanghai AI Laboratory（上海人工智能实验室） ； MOE KLINNS Lab Xi’an Jiaotong University China（西安交通大学教育部KLINNS实验室） 纠错

专题命中 安全评测 ：safety(title,abstract)；alignment(abstract)；分类 cs.CL

Enrich-on-Graph: Query-Graph Alignment for Complex Reasoning with LLM Enriching

Songze Li, Zhiqiang Liu, Zhengke Gui, Huajun Chen, Wen Zhang

机构 * Zhejiang University（浙江大学） ； Ant Group（蚂蚁集团） ； ZJU-Ant Group Joint Lab of Knowledge Graph（浙江大学-蚂蚁集团知识图谱联合实验室） 纠错

专题命中 安全评测 ：alignment(title,abstract)；分类 cs.CL

Journal ref EMNLP 2025, pages 7683-7703

Bridged Semantic Alignment for Zero-shot 3D Medical Image Diagnosis

Haoran Lai, Zihang Jiang, Qingsong Yao, Rongsheng Wang, Zhiyang He, Xiaodong Tao, Weifu Lv, Wei Wei, S. Kevin Zhou

机构 * University of Science and Technology of China（中国科学技术大学） ； Suzhou Institute for Advanced Research（苏州先进研究所） ； Stanford University（斯坦福大学） ； iFlytek Co. Ltd.（iFlytek公司） ； The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, USTC（中国科学技术大学第一附属医院） 纠错

专题命中 安全评测 ：alignment(title,abstract)

Selection of LLM Fine-Tuning Data based on Orthogonal Rules

Xiaomin Li, Mingye Gao, Zhiwei Zhang, Chang Yue, Hong Hu

专题命中 安全评测 ：alignment(abstract)；分类 cs.CL、cs.AI、cs.LG

NOTAM-Evolve: A Knowledge-Guided Self-Evolving Optimization Framework with LLMs for NOTAM Interpretation

Maoqi Liu, Quan Fang, Yuhao Wu, Can Zhao, Yang Yang, Kaiquan Cai

专题命中 安全评测 ：safety(abstract)；分类 cs.CL、cs.AI

Comments Accepted to AAAI 2026

Revisiting NLI: Towards Cost-Effective and Human-Aligned Metrics for Evaluating LLMs in Question Answering

Sai Shridhar Balamurali, Lu Cheng

机构 * University of Illinois at Chicago（伊利诺伊大学香槟分校） 纠错

专题命中 安全评测 ：alignment(abstract)；分类 cs.CL、cs.AI

Revisiting Network Traffic Analysis: Compatible network flows for ML models

João Vitorino, Daniela Pinto, Eva Maia, Ivone Amorim, Isabel Praça

机构 * GECAD, ISEP, Polytechnic of Porto（GECAD、ISEP、波尔图理工大学） 纠错

专题命中 安全评测 ：trustworthy(abstract)；分类 cs.LG

Comments 16 pages, 12 tables, 1 figure, FPS 2025 conference

Beyond the Pixels: VLM-based Evaluation of Identity Preservation in Reference-Guided Synthesis

Aditi Singhania, Krutik Malani, Riddhi Dhawan, Arushi Jain, Garv Tandon, Nippun Sharma, Souymodip Chakraborty, Vineet Batra, Ankit Phogat

专题命中 安全评测 ：alignment(abstract)；分类 cs.AI