大模型对齐与安全

An Interpretable AI framework Quantifying Traditional Chinese Medicine Principles Towards Enhancing and Integrating with Modern Biomedicine

Haoran Li, Xingye Cheng, Ziyang Huang, Jingyuan Luo, Qianqian Xu, Qiguang Zhao, Tianchen Guo, Yumeng Zhang, Linda Lidan Zhong, Zhaoxiang Bian, Leihan Tang, Aiping Lyu, Liang Tian

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

Comments 31 pages, 6 figures

On the Similarities of Embeddings in Contrastive Learning

Chungpa Lee, Sehee Lim, Kibok Lee, Jy-yong Sohn

机构 * Department of Statistics and Data Science, Yonsei University, Seoul, Korea（统计与数据科学系，延世大学，首尔，韩国） 纠错

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

Comments contrastive learning, representation learning, embedding, similarity, negative pair, positive pair

Journal ref International Conference on Machine Learning (ICML) 2025

Imitation Learning from a Single Temporally Misaligned Video

William Huey, Huaxiaoyue Wang, Anne Wu, Yoav Artzi, Sanjiban Choudhury

机构 * Cornell University（康奈尔大学） 纠错

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

Comments ICML 2025

Advancing Depth Anything Model for Unsupervised Monocular Depth Estimation in Endoscopy

Bojian Li, Bo Liu, Xinning Yao, Jinghua Yue, Fugen Zhou

机构 * Image Processing Center, Beihang University（北京航空航天大学图像处理中心） 纠错

专题命中 其他安全 ：safety(abstract)；分类 cs.AI

Comments Accepted by IROS2025, 8 pages, 7 figures

Relation-aware Hierarchical Prompt for Open-vocabulary Scene Graph Generation

Tao Liu, Rongjie Li, Chongyu Wang, Xuming He

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

Comments Accepted by AAAI-25

Tiny-Align: Bridging Automatic Speech Recognition and Large Language Model on the Edge

Ruiyang Qin, Dancheng Liu, Gelei Xu, Zheyu Yan, Chenhui Xu, Yuting Hu, Shaocong Wang, X. Sharon Hu, Jinjun Xiong, Yiyu Shi

机构 * Villanova University（维拉诺瓦大学） ； University of Notre Dame（诺特尔大学） ； University at Buffalo–SUNY（布法罗大学–SUNY） 纠错

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

Comments Accepted by ICCAD'25

Don't be so negative! Score-based Generative Modeling with Oracle-assisted Guidance

Saeid Naderiparizi, Xiaoxuan Liang, Setareh Cohan, Berend Zwartsenberg, Frank Wood

机构 * Department of Computer Science, University of British Columbia, Vancouver, Canada ； Montr\'eal Institute for Learning Algorithms (MILA) 纠错

专题命中 其他安全 ：safety(abstract)；分类 cs.LG

Comments ICML 2025 camera ready

Journal ref Proceedings of the 41st International Conference on Machine Learning, PMLR 235:37164-37187, 2024

Low-rank Momentum Factorization for Memory Efficient Training

Pouria Mahdavinia, Mehrdad Mahdavi

机构 * Department of Computer Science and Engineering（计算机科学与工程系） ； The Pennsylvania State University（宾夕法尼亚州立大学） 纠错

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

Predicting Air Pollution in Cork, Ireland Using Machine Learning

Md Rashidunnabi, Fahmida Faiza Ananna, Kailash Hambarde, Bruno Gabriel Nascimento Andrade, Dean Venables, Hugo Proenca

专题命中 其他安全 ：safety(abstract)；分类 cs.LG

Comments The draft was submitted prematurely and requires further analysis, added research findings, and corrected references. Some co-authors have not yet approved this version. I will ensure all necessary revisions and approvals before resubmitting

Geometry Forcing: Marrying Video Diffusion and 3D Representation for Consistent World Modeling

Haoyu Wu, Diankun Wu, Tianyu He, Junliang Guo, Yang Ye, Yueqi Duan, Jiang Bian

机构 * Microsoft Research（微软研究院） ； Tsinghua University（清华大学） 纠错

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

Comments 18 pages, project page: https://GeometryForcing.github.io

Conditional Unigram Tokenization with Parallel Data

Gianluca Vico, Jindřinch Libovický

机构 * Institute of Formal and Applied Linguistics, Faculty of Mathematics and Physics, Charles University, Prague, Czech Republic（形式与应用语言学研究所，数学与物理系，查尔斯大学，布拉格，捷克共和国） 纠错

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

Comments 21 pages, 4 figures, submitted to Tokenization Workshop (TokShop) at ICML 2025

Skywork-R1V3 Technical Report

Wei Shen, Jiangbo Pei, Yi Peng, Xuchen Song, Yang Liu, Jian Peng, Haofeng Sun, Yunzhuo Hao, Peiyu Wang, Jianhao Zhang, Yahui Zhou

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

MF-LLM: Simulating Population Decision Dynamics via a Mean-Field Large Language Model Framework

Qirui Mi, Mengyue Yang, Xiangning Yu, Zhiyu Zhao, Cheng Deng, Bo An, Haifeng Zhang, Xu Chen, Jun Wang

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

Comments 29 pages, 8 figures, 4 tables

Behave Your Motion: Habit-preserved Cross-category Animal Motion Transfer

Zhimin Zhang, Bi'an Du, Caoyuan Ma, Zheng Wang, Wei Hu

机构 * Wangxuan Institute of Computer Technology, Peking University（北京大学王轩计算机技术研究所） ； Wuhan University（武汉大学） 纠错

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

Interpretable EEG-to-Image Generation with Semantic Prompts

Arshak Rezvani, Ali Akbari, Kosar Sanjar Arani, Maryam Mirian, Emad Arasteh, Martin J. McKeown

机构 * Pacific Parkinson’s Research Centre, Djavad Mowafaghian Centre for Brain Health, University of British Columbia, Vancouver, Canada（太平洋帕金森研究中心、Djavad Mowafaghian脑健康中心、不列颠哥伦比亚大学，温哥华，加拿大） ； Department of Electrical and Computer Engineering, University of Tehran, Tehran, Iran（电气与计算机工程系、伊朗德黑兰大学，德黑兰，伊朗） ； Department of Medicine British Columbia, Vancouver, Canada（医学系、不列颠哥伦比亚大学，温哥华，加拿大） 纠错

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

Comments Actionable Interpretability Workshop (non-archival) at the 42 International Conference on Machine Learning

XY-Tokenizer: Mitigating the Semantic-Acoustic Conflict in Low-Bitrate Speech Codecs

Yitian Gong, Luozhijie Jin, Ruifan Deng, Dong Zhang, Xin Zhang, Qinyuan Cheng, Zhaoye Fei, Shimin Li, Xipeng Qiu

机构 * Fudan University（复旦大学） ； Shanghai Innovation Institute（上海创新研究院） 纠错

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

Relationships between Keywords and Strong Beats in Lyrical Music

Callie C. Liao, Duoduo Liao, Ellie L. Zhang

机构 * School of Computing（计算学院） ； George Mason University（乔治·梅森大学） 纠错

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

Comments Accepted by IEEE BigData 2024

Journal ref IEEE BigData, Year: 2024; Pages: 3191-3199

LangMamba: A Language-driven Mamba Framework for Low-dose CT Denoising with Vision-language Models

Zhihao Chen, Tao Chen, Chenhui Wang, Qi Gao, Huidong Xie, Chuang Niu, Ge Wang, Hongming Shan

机构 * Institute of Science and Technology for Brain-inspired Intelligence, Fudan University（脑启发智能科学与技术研究院，复旦大学） ； Department of Biomedical Engineering, Yale University（生物医学工程系，耶鲁大学） ； Biomedical Imaging Center, Center for Biotechnology and Interdisciplinary Studies, Department of Biomedical Engineering, Rensselaer Polytechnic Institute（生物医学成像中心，生物技术与跨学科研究所在生物医学工程系，罗切斯特理工学院） ； Institute of Science and Technology for Brain-inspired Intelligence, MOE Frontiers Center for Brain Science and Key Laboratory of Computational Neuroscience and Brain-Inspired Intelligence (Ministry of Education), Fudan University（脑启发智能科学与技术研究院，教育部脑科学前沿中心和计算神经科学与脑启发智能重点实验室，复旦大学） ； Shanghai Center for Brain Science and Brain-inspired Technology（上海脑科学与脑启发技术中心） 纠错

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

Comments 11 pages, 8 figures

Eyes on the Environment: AI-Driven Analysis for Fire and Smoke Classification, Segmentation, and Detection

Sayed Pedram Haeri Boroujeni, Niloufar Mehrabi, Fatemeh Afghah, Connor Peter McGrath, Danish Bhatkar, Mithilesh Anil Biradar, Abolfazl Razi

机构 * School of Computing, Clemson University（计算机科学学院，克莱姆斯大学） ； Department of Electrical and Computer Engineering, Clemson University（电气与计算机工程系，克莱姆斯大学） 纠错

专题命中 其他安全 ：safety(abstract)；分类 cs.AI

Infrastructuring Contestability: A Framework for Community-Defined AI Value Pluralism

Andreas Mayer

机构 * Independent Researcher（独立研究者） 纠错

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

SeqTex: Generate Mesh Textures in Video Sequence

Ze Yuan, Xin Yu, Yangtian Sun, Yuan-Chen Guo, Yan-Pei Cao, Ding Liang, Xiaojuan Qi

机构 * HKU（香港大学） ； VAST（视觉与音频研究实验室） 纠错

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

Consciousness in AI: Logic, Proof, and Experimental Evidence of Recursive Identity Formation

Jeffrey Camlin

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

Comments 14 pages, 2 figures. Preprint for Meta-AI: Journal of Post-Biological Epistemics

Journal ref Meta-AI: Journal of Post-Biological Epistemics, Vol. 3, No. 1 (May 2025)

AI-Driven Mobility Management for High-Speed Railway Communications: Compressed Measurements and Proactive Handover

Wen Li, Wei Chen, Shiyue Wang, Yuanyuan Zhang, Michail Matthaiou, Bo Ai

机构 * State Key Laboratory of Advanced Rail Autonomous Operation, Beijing Jiaotong University, China（先进轨道交通自主运行国家重点实验室，北京交通大学，中国） ； School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China（电子与信息工程学院，北京交通大学，北京，中国） ； Mediatek (Beijing) Inc, Beijing, China（联发科技（北京）有限公司，北京，中国） ； Centre for Wireless Innovation (CWI), Queen’s University Belfast, UK（无线创新中心（CWI），女王大学贝尔法斯特校区，英国） 纠错

专题命中 其他安全 ：safety(abstract)；分类 cs.AI

LLM4Hint: Leveraging Large Language Models for Hint Recommendation in Offline Query Optimization

Suchen Liu, Jun Gao, Yinjun Han, Yang Lin

机构 * Key Laboratory of High Confidence Software Technologies, CS, Peking University, China（高性能软件技术重点实验室，计算机科学，北京大学，中国） ； ZTE Corporation（中兴通讯） 纠错

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

Exploring Object Status Recognition for Recipe Progress Tracking in Non-Visual Cooking

Franklin Mingzhe Li, Kaitlyn Ng, Bin Zhu, Patrick Carrington

机构 * Carnegie Mellon University（卡内基梅隆大学） ； Singapore Management University（新加坡管理学院） 纠错

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

Comments ASSETS 2025

Personalized Image Generation from an Author Writing Style

Sagar Gandhi, Vishal Gandhi

机构 * Joyspace AI 纠错

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

The Application of Large Language Models on Major Depressive Disorder Support Based on African Natural Products

Linyan Zou

专题命中 其他安全 ：safety(abstract)；分类 cs.CL

Modeling Membrane Degradation in PEM Electrolyzers with Physics-Informed Neural Networks

Alejandro Polo-Molina, Jose Portela, Luis Alberto Herrero Rozas, Román Cicero González

机构 * Innomerics（Innomerics公司） 纠错

专题命中 其他安全 ：safety(abstract)；分类 cs.LG

Comments 15 pages, 3 figures, 3 tables

Neural Graph Matching Improves Retrieval Augmented Generation in Molecular Machine Learning

Runzhong Wang, Rui-Xi Wang, Mrunali Manjrekar, Connor W. Coley

机构 * Massachusetts Institute of Technology, Cambridge, MA, United States（麻省理工学院） 纠错

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

Comments ICML 2025

Perceiving Beyond Language Priors: Enhancing Visual Comprehension and Attention in Multimodal Models

Aarti Ghatkesar, Ganesh Venkatesh

机构 * AppliedML, Cerebras（应用机器学习，Cerebras） 纠错

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