大模型对齐与安全 - arXivDaily 专题

2510.18154 2025-10-22 cs.AI cs.CY 88%

Annotating the Chain-of-Thought: A Behavior-Labeled Dataset for AI Safety

Antonio-Gabriel Chacón Menke, Phan Xuan Tan, Eiji Kamioka

专题命中其他安全：safety(title,abstract)；AI safety(title,abstract)；分类 cs.AI、cs.CY

2507.15886 2025-10-22 cs.CY cs.AI 87%

Combining Cost-Constrained Runtime Monitors for AI Safety

Tim Tian Hua, James Baskerville, Henri Lemoine, Mia Hopman, Aryan Bhatt, Tyler Tracy

机构 * MARS ； Redwood Research

专题命中其他安全：safety(title,abstract)；AI safety(title)；分类 cs.AI、cs.CY

2510.17833 2025-10-22 q-bio.NC cs.AI 79%

Brain-Language Model Alignment: Insights into the Platonic Hypothesis and Intermediate-Layer Advantage

Ángela López-Cardona, Sebastián Idesis, Mireia Masias-Bruns, Sergi Abadal, Ioannis Arapakis

机构 * Universitat Politècnica de Catalunya（加泰罗尼亚理工大学）

专题命中其他安全：alignment(title,abstract)；分类 cs.AI

2505.14359 2025-10-22 cs.CV 78%

Dual Data Alignment Makes AI-Generated Image Detector Easier Generalizable

Ruoxin Chen, Junwei Xi, Zhiyuan Yan, Ke-Yue Zhang, Shuang Wu, Jingyi Xie, Xu Chen, Lei Xu, Isabel Guan, Taiping Yao, Shouhong Ding

机构 * Tencent YouTu Lab（腾讯优图实验室）； East China University of Science and Technology（东华大学）； Peking University（北京大学）； Renmin University of China（中国人民大学）； Shenzhen University（深圳大学）； Hong Kong University of Science and Technology（香港科技大学）

专题命中其他安全：alignment(title,abstract)

Comments NeurIPS 2025 Spotlight. 13 Pages, 10 figures

2510.18502 2025-10-22 cs.CV cs.AI cs.CL cs.LG 67%

Zero-Shot Vehicle Model Recognition via Text-Based Retrieval-Augmented Generation

Wei-Chia Chang, Yan-Ann Chen

机构 * Yuan Ze University（元智大学）

专题命中其他安全：alignment(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Accepted by The 38th Conference of Open Innovations Association FRUCT, 2025

2502.01472 2025-10-22 cs.CL cs.AI 62%

FALCON: Fine-grained Activation Manipulation by Contrastive Orthogonal Unalignment for Large Language Model

Jinwei Hu, Zhenglin Huang, Xiangyu Yin, Wenjie Ruan, Guangliang Cheng, Yi Dong, Xiaowei Huang

机构 * School of Computer Science and Informatics, University of Liverpool, UK（计算机科学与信息学学院，利物浦大学）

专题命中其他安全：safety(abstract)；分类 cs.CL、cs.AI

Comments Accepted at NeurIPS 2025 with minor revisions

2510.18745 2025-10-22 cs.CL 61%

Topoformer: brain-like topographic organization in Transformer language models through spatial querying and reweighting

Taha Binhuraib, Greta Tuckute, Nicholas Blauch

机构 * Novus Technologies ； MIT（麻省理工学院）； Harvard University（哈佛大学）

专题命中其他安全：alignment(abstract,comments)；分类 cs.CL

Comments ICLR 2024 Workshop on Representational Alignment (Re-Align) Camera Ready

2510.18583 2025-10-22 cs.CV cs.LG 57%

CovMatch: Cross-Covariance Guided Multimodal Dataset Distillation with Trainable Text Encoder

Yongmin Lee, Hye Won Chung

专题命中其他安全：alignment(abstract)；分类 cs.LG

Comments NeurIPS 2025

2506.00363 2025-10-22 cs.IR cs.CL 57%

Adapting General-Purpose Embedding Models to Private Datasets Using Keyword-based Retrieval

Yubai Wei, Jiale Han, Yi Yang

专题命中其他安全：alignment(abstract)；分类 cs.CL

Comments Link: https://github.com/BaileyWei/BMEmbed

Journal ref Findings of the Association for Computational Linguistics ACL 2025

2510.17909 2025-10-22 cs.CL 57%

Atomic Literary Styling: Mechanistic Manipulation of Prose Generation in Neural Language Models

Tsogt-Ochir Enkhbayar

机构 * Mongol AI（蒙古AI）

专题命中其他安全：alignment(abstract)；分类 cs.CL

Comments 12 pages, 3 figures, 4 tables

2510.18113 2025-10-22 cs.CR 50%

Investigating the Impact of Dark Patterns on LLM-Based Web Agents

Devin Ersoy, Brandon Lee, Ananth Shreekumar, Arjun Arunasalam, Muhammad Ibrahim, Antonio Bianchi, Z. Berkay Celik

专题命中其他安全：safety(abstract)

Comments At IEEE S&P 2026

1. 其他安全 11 篇