大模型对齐与安全

Integrating Chain-of-Thought for Multimodal Alignment: A Study on 3D Vision-Language Learning

Yanjun Chen, Yirong Sun, Xinghao Chen, Jian Wang, Xiaoyu Shen, Wenjie Li, Wei Zhang

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

A Survey on Trustworthy LLM Agents: Threats and Countermeasures

Miao Yu, Fanci Meng, Xinyun Zhou, Shilong Wang, Junyuan Mao, Linsey Pang, Tianlong Chen, Kun Wang, Xinfeng Li, Yongfeng Zhang, Bo An, Qingsong Wen

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

Architecture for a Trustworthy Quantum Chatbot

Yaiza Aragonés-Soria, Manuel Oriol

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

Know Thy Judge: On the Robustness Meta-Evaluation of LLM Safety Judges

Francisco Eiras, Eliott Zemour, Eric Lin, Vaikkunth Mugunthan

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

Comments Accepted to the ICBINB Workshop at ICLR'25

CoT2Align: Cross-Chain of Thought Distillation via Optimal Transport Alignment for Language Models with Different Tokenizers

Anh Duc Le, Tu Vu, Nam Le Hai, Nguyen Thi Ngoc Diep, Linh Ngo Van, Trung Le, Thien Huu Nguyen

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

Conformal Tail Risk Control for Large Language Model Alignment

Catherine Yu-Chi Chen, Jingyan Shen, Zhun Deng, Lihua Lei

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

Trustworthy AI-Generative Content for Intelligent Network Service: Robustness, Security, and Fairness

Siyuan Li, Xi Lin, Yaju Liu, Xiang Chen, Jianhua Li

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

Coherence-Driven Multimodal Safety Dialogue with Active Learning for Embodied Agents

Sabit Hassan, Hye-Young Chung, Xiang Zhi Tan, Malihe Alikhani

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

Comments To appear at AAMAS, 2025

Benchmarking Temporal Reasoning and Alignment Across Chinese Dynasties

Zhenglin Wang, Jialong Wu, Pengfei LI, Yong Jiang, Deyu Zhou

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

Comments Preprint

Adversarial Alignment for LLMs Requires Simpler, Reproducible, and More Measurable Objectives

Leo Schwinn, Yan Scholten, Tom Wollschläger, Sophie Xhonneux, Stephen Casper, Stephan Günnemann, Gauthier Gidel

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

Evaluation of Best-of-N Sampling Strategies for Language Model Alignment

Yuki Ichihara, Yuu Jinnai, Tetsuro Morimura, Kaito Ariu, Kenshi Abe, Mitsuki Sakamoto, Eiji Uchibe

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

Journal ref Transactions on Machine Learning Research (TMLR), 2025

AgentGuard: Repurposing Agentic Orchestrator for Safety Evaluation of Tool Orchestration

Jizhou Chen, Samuel Lee Cong

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

Comments Project report of AgentGuard in LLM Agent MOOC Hackathon hosted by UC Berkeley in 2024

Efficient and Trustworthy Block Propagation for Blockchain-enabled Mobile Embodied AI Networks: A Graph Resfusion Approach

Jiawen Kang, Jiana Liao, Runquan Gao, Jinbo Wen, Huawei Huang, Maomao Zhang, Changyan Yi, Tao Zhang, Dusit Niyato, Zibin Zheng

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

Comments 15 pages, 11 figures

Advantage Alignment Algorithms

Juan Agustin Duque, Milad Aghajohari, Tim Cooijmans, Razvan Ciuca, Tianyu Zhang, Gauthier Gidel, Aaron Courville

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

Comments 25 Pages, 8 figures

Explainable and Robust Millimeter Wave Beam Alignment for AI-Native 6G Networks

Nasir Khan, Asmaa Abdallah, Abdulkadir Celik, Ahmed M. Eltawil, Sinem Coleri

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

Early External Safety Testing of OpenAI's o3-mini: Insights from the Pre-Deployment Evaluation

Aitor Arrieta, Miriam Ugarte, Pablo Valle, José Antonio Parejo, Sergio Segura

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

Comments arXiv admin note: text overlap with arXiv:2501.17132

CHG Shapley: Efficient Data Valuation and Selection towards Trustworthy Machine Learning

Huaiguang Cai

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

Comments Rejected by ICLR 2025. https://openreview.net/forum?id=uVMZgtw2pf

ReARTeR: Retrieval-Augmented Reasoning with Trustworthy Process Rewarding

Zhongxiang Sun, Qipeng Wang, Weijie Yu, Xiaoxue Zang, Kai Zheng, Jun Xu, Xiao Zhang, Song Yang, Han Li

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

Comments 11 pages, 5 figures

Hierarchical Divide-and-Conquer for Fine-Grained Alignment in LLM-Based Medical Evaluation

Shunfan Zheng, Xiechi Zhang, Gerard de Melo, Xiaoling Wang, Linlin Wang

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

CoDe: Communication Delay-Tolerant Multi-Agent Collaboration via Dual Alignment of Intent and Timeliness

Shoucheng Song, Youfang Lin, Sheng Han, Chang Yao, Hao Wu, Shuo Wang, Kai Lv

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

Comments AAAI 2025 Accepted

Pathway to Secure and Trustworthy ZSM for LLMs: Attacks, Defense, and Opportunities

Sunder Ali Khowaja, Parus Khuwaja, Kapal Dev, Hussam Al Hamadi, Engin Zeydan

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

Comments 7 pages, 4 figures

Feature Alignment-Based Knowledge Distillation for Efficient Compression of Large Language Models

Shuo Wang, Chihang Wang, Jia Gao, Zhen Qi, Hongye Zheng, Xiaoxuan Liao

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

Comments 4 pages

Context-Based Semantic-Aware Alignment for Semi-Supervised Multi-Label Learning

Heng-Bo Fan, Ming-Kun Xie, Jia-Hao Xiao, Sheng-Jun Huang

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

Agents on the Bench: Large Language Model Based Multi Agent Framework for Trustworthy Digital Justice

Cong Jiang, Xiaolei Yang

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

Comments Draft version; Under review

Trustworthy and Efficient LLMs Meet Databases

Kyoungmin Kim, Anastasia Ailamaki

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

Python Fuzzing for Trustworthy Machine Learning Frameworks

Ilya Yegorov, Eli Kobrin, Darya Parygina, Alexey Vishnyakov, Andrey Fedotov

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

Journal ref Journal of Mathematical Sciences, 2024 Springer Nature Switzerland AG, Vol. 285, No. 2, October, 2024

FFT: Towards Harmlessness Evaluation and Analysis for LLMs with Factuality, Fairness, Toxicity

Shiyao Cui, Zhenyu Zhang, Yilong Chen, Wenyuan Zhang, Tianyun Liu, Siqi Wang, Tingwen Liu

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

Comments Accepted by KDD workshop on Evaluation and Trustworthiness of Generative AI Models

Journal ref https://genai-evaluation-kdd2024.github.io/genai-evalution-kdd2024/assets/papers/GenAI_Evaluation_KDD2024_paper_5.pdf

On Robust Cross Domain Alignment

Anish Chakrabarty, Arkaprabha Basu, Swagatam Das

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

I-SHEEP: Self-Alignment of LLM from Scratch through an Iterative Self-Enhancement Paradigm

Yiming Liang, Ge Zhang, Xingwei Qu, Tianyu Zheng, Jiawei Guo, Xinrun Du, Zhenzhu Yang, Jiaheng Liu, Chenghua Lin, Lei Ma, Wenhao Huang, Jiajun Zhang

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

Bridging Sequence-Structure Alignment in RNA Foundation Models

Heng Yang, Renzhi Chen, Ke Li

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

Comments Accepted by AAAI 2025