大模型对齐与安全

Escaping Plato's Cave: Towards the Alignment of 3D and Text Latent Spaces

Souhail Hadgi, Luca Moschella, Andrea Santilli, Diego Gomez, Qixing Huang, Emanuele Rodolà, Simone Melzi, Maks Ovsjanikov

机构 * École polytechnique（巴黎政治学院） ； Sapienza University of Rome（罗马大学） ； University of Milano-Bicocca（米兰-布雷拉大学） ； The University of Texas at Austin（德克萨斯大学奥斯汀分校） 纠错

专题命中 其他安全 ：alignment(title,abstract)

Comments CVPR 2025

Kernel-based Unsupervised Embedding Alignment for Enhanced Visual Representation in Vision-language Models

Shizhan Gong, Yankai Jiang, Qi Dou, Farzan Farnia

专题命中 其他安全 ：alignment(title,abstract)

Comments ICML 2025

Urban Safety Perception Assessments via Integrating Multimodal Large Language Models with Street View Images

Jiaxin Zhang, Yunqin Li, Tomohiro Fukuda, Bowen Wang

专题命中 其他安全 ：safety(title,abstract)

Comments 15 pages, 10 figures

VideoREPA: Learning Physics for Video Generation through Relational Alignment with Foundation Models

Xiangdong Zhang, Jiaqi Liao, Shaofeng Zhang, Fanqing Meng, Xiangpeng Wan, Junchi Yan, Yu Cheng

机构 * Dept. of CSE & School of AI & MoE Key Lab of Al, Shanghai Jiao Tong University（计算机科学与工程系及人工智能学院及人工智能基础实验室，上海交通大学） ； Shanghai Jiao Tong University（上海交通大学） ； The Chinese University of Hong Kong（香港中文大学） 纠错

专题命中 其他安全 ：alignment(title,abstract)

User Feedback Alignment for LLM-powered Exploration in Large-scale Recommendation Systems

Jianling Wang, Yifan Liu, Yinghao Sun, Xuejian Ma, Yueqi Wang, He Ma, Zhengyang Su, Minmin Chen, Mingyan Gao, Onkar Dalal, Ed H. Chi, Lichan Hong, Ningren Han, Haokai Lu

专题命中 其他安全 ：alignment(title,abstract)

Comments ACL'25 (Industry) Oral

A Unifying Perspective for Safety of Stochastic Systems: From Barrier Functions to Finite Abstractions

Luca Laurenti, Morteza Lahijanian

专题命中 其他安全 ：safety(title,abstract)

Comments Submitted to IEEE Transaction on Automatic Control

Focus on What Matters: Enhancing Medical Vision-Language Models with Automatic Attention Alignment Tuning

Aofei Chang, Le Huang, Alex James Boyd, Parminder Bhatia, Taha Kass-Hout, Cao Xiao, Fenglong Ma

专题命中 其他安全 ：alignment(title,abstract)

Comments Accepted to ACL2025 (main)

Beyond General Prompts: Automated Prompt Refinement using Contrastive Class Alignment Scores for Disambiguating Objects in Vision-Language Models

Lucas Choi, Ross Greer

机构 * University of California, Merced（加州大学默塞德分校） 纠错

专题命中 其他安全 ：alignment(title,abstract)

Anatomical Attention Alignment representation for Radiology Report Generation

Quang Vinh Nguyen, Minh Duc Nguyen, Thanh Hoang Son Vo, Hyung-Jeong Yang, Soo-Hyung Kim

专题命中 其他安全 ：alignment(title,abstract)

TeDA: Boosting Vision-Lanuage Models for Zero-Shot 3D Object Retrieval via Testing-time Distribution Alignment

Zhichuan Wang, Yang Zhou, Jinhai Xiang, Yulong Wang, Xinwei He

机构 * Huazhong Agricultural University（华中农业大学） 纠错

专题命中 其他安全 ：alignment(title,abstract)

Comments Accepted by ICMR 2025

Text-to-Image Alignment in Denoising-Based Models through Step Selection

Paul Grimal, Hervé Le Borgne, Olivier Ferret

机构 * Université Paris-Saclay, CEA, List（巴黎-萨克雷大学、法国原子能委员会、List） 纠错

专题命中 其他安全 ：alignment(title,abstract)

LLMCode: Evaluating and Enhancing Researcher-AI Alignment in Qualitative Analysis

Joel Oksanen, Andrés Lucero, Perttu Hämäläinen

专题命中 其他安全 ：alignment(title,abstract)

Assessing and Learning Alignment of Unimodal Vision and Language Models

Le Zhang, Qian Yang, Aishwarya Agrawal

机构 * Mila - Quebec AI Institute（魁北克人工智能研究所） ； Université de Montréal（蒙特利尔大学） ； Canada CIFAR AI Chair（加拿大CIFAR人工智能 chair） 纠错

专题命中 其他安全 ：alignment(title,abstract)

Comments CVPR 2025 Highlight

On the Alignment of Post-Publication Reviews & Bibliometric and Altmetric Impact -- A Case Study on Expert Statements from the Science Media Center Germany

Dirk Tunger, Philipp Schaer

专题命中 其他安全 ：alignment(title,abstract)

Comments Accepted at The First Workshop on Scholarly Information Access (SCOLIA)

The Role of Computational Modeling in Enhancing Thermal Safety During Cardiac Ablation

Leila Seidabadi, Indra Vandenbussche, Rowan Carter Fink, MacKenzie Moore, Bailey McCorkendale, Fateme Esmailie

专题命中 其他安全 ：safety(title,abstract)

Comments The Manuscript was submitted for review to ICVTS Journal, 40 pages, 1 figure, and 3 tables. Fateme Esmailie is the corresponding author. Leila Seidabadi and Indra Vandenbussche are Co-first authors

GOAL: Global-local Object Alignment Learning

Hyungyu Choi, Young Kyun Jang, Chanho Eom

专题命中 其他安全 ：alignment(title,abstract)

Comments 16 pages, 5 figures

Can Text-to-Video Generation help Video-Language Alignment?

Luca Zanella, Massimiliano Mancini, Willi Menapace, Sergey Tulyakov, Yiming Wang, Elisa Ricci

专题命中 其他安全 ：alignment(title,abstract)

Comments CVPR 2025. Project website at https://lucazanella.github.io/synvita/

Harnessing Frozen Unimodal Encoders for Flexible Multimodal Alignment

Mayug Maniparambil, Raiymbek Akshulakov, Yasser Abdelaziz Dahou Djilali, Sanath Narayan, Ankit Singh, Noel E. O'Connor

专题命中 其他安全 ：alignment(title,abstract)

Comments Accepted CVPR 2025; First two authors contributed equally;

Safety for Time-Varying Parameterized Sets Using Control Barrier Function Methods

James Usevitch, Jackson Sahleen

专题命中 其他安全 ：safety(title,abstract)

HALO: Fault-Tolerant Safety Architecture For High-Speed Autonomous Racing

Aron Harder, Amar Kulkarni, Madhur Behl

专题命中 其他安全 ：safety(title,abstract)

Comments 27 pages, 7 figures

Towards Improved Text-Aligned Codebook Learning: Multi-Hierarchical Codebook-Text Alignment with Long Text

Guotao Liang, Baoquan Zhang, Zhiyuan Wen, Junteng Zhao, Yunming Ye, Kola Ye, Yao He

专题命中 其他安全 ：alignment(title,abstract)

Comments Accepted by CVPR 2025

Hierarchical Cross-Modal Alignment for Open-Vocabulary 3D Object Detection

Youjun Zhao, Jiaying Lin, Rynson W. H. Lau

专题命中 其他安全 ：alignment(title,abstract)

Comments AAAI 2025 (Extented Version). Project Page: https://youjunzhao.github.io/HCMA/

OV-SCAN: Semantically Consistent Alignment for Novel Object Discovery in Open-Vocabulary 3D Object Detection

Adrian Chow, Evelien Riddell, Yimu Wang, Sean Sedwards, Krzysztof Czarnecki

专题命中 其他安全 ：alignment(title,abstract)

DoPTA: Improving Document Layout Analysis using Patch-Text Alignment

Nikitha SR, Tarun Ram Menta, Mausoom Sarkar

专题命中 其他安全 ：alignment(title,abstract)

ScVLM: Enhancing Vision-Language Model for Safety-Critical Event Understanding

Liang Shi, Boyu Jiang, Tong Zeng, Feng Guo

专题命中 其他安全 ：safety(title,abstract)

Comments To appear in Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2025

Journal ref Proceedings of the Winter Conference on Applications of Computer Vision (WACV) Workshops, 2025, pp. 1061-1071

Semantic Alignment of Unimodal Medical Text and Vision Representations

Maxime Di Folco, Emily Chan, Marta Hasny, Cosmin I. Bercea, Julia A. Schnabel

专题命中 其他安全 ：alignment(title,abstract)

M2Lingual: Enhancing Multilingual, Multi-Turn Instruction Alignment in Large Language Models

Rishabh Maheshwary, Vikas Yadav, Hoang Nguyen, Khyati Mahajan, Sathwik Tejaswi Madhusudhan

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

Comments 39 pages

MFM-DA: Instance-Aware Adaptor and Hierarchical Alignment for Efficient Domain Adaptation in Medical Foundation Models

Jia-Xuan Jiang, Wenhui Lei, Yifeng Wu, Hongtao Wu, Furong Li, Yining Xie, Xiaofan Zhang, Zhong Wang

专题命中 其他安全 ：alignment(title,abstract)

A RISC-V Multicore and GPU SoC Platform with a Qualifiable Software Stack for Safety Critical Systems

Marc Solé i Bonet, Jannis Wolf, Leonidas Kosmidis

专题命中 其他安全 ：safety(title,abstract)

Comments OSSMPIC2025, 1st workshop on Open Source Solutions for Massively Parallel Integrated Circuits

User-Driven Value Alignment: Understanding Users' Perceptions and Strategies for Addressing Biased and Discriminatory Statements in AI Companions

Xianzhe Fan, Qing Xiao, Xuhui Zhou, Jiaxin Pei, Maarten Sap, Zhicong Lu, Hong Shen

专题命中 其他安全 ：alignment(title,abstract)

Comments 19 pages, 1 figure, Accepted by CHI '25