大模型对齐与安全

Gaussian process surrogate model to approximate power grid simulators -- An application to the certification of a congestion management controller

Pierre Houdouin, Lucas Saludjian

机构 * RTE PARIS, FRANCE（法国巴黎电力公司） 纠错

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

Comments 6 pages, 7 figures, conference

Span-Level Hallucination Detection for LLM-Generated Answers

Passant Elchafei, Mervet Abu-Elkheir

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

A comprehensive review of classifier probability calibration metrics

Richard Oliver Lane

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

Comments 60 pages, 7 figures

Smooth Calibration and Decision Making

Jason Hartline, Yifan Wu, Yunran Yang

机构 * Northwestern University（西北大学） ； Shanghai Jiaotong University（上海交通大学） 纠错

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

Comments In FORC 2025

A Genetic Fuzzy-Enabled Framework on Robotic Manipulation for In-Space Servicing

Nathan Steffen, Wilhelm Louw, Nicholas Ernest, Timothy Arnett, Kelly Cohen

机构 * University of Cincinnati（克利夫兰大学） ； Thales Avionics（泰勒斯航空电子） 纠错

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

Object-Level Verbalized Confidence Calibration in Vision-Language Models via Semantic Perturbation

Yunpu Zhao, Rui Zhang, Junbin Xiao, Ruibo Hou, Jiaming Guo, Zihao Zhang, Yifan Hao, Yunji Chen

机构 * University of Science and Technology of China（中国科学技术大学） ； SKL of Processors, Institute of Computing Technology, CAS（中国科学院计算技术研究所处理器专项实验室） ； National University of Singapore（新加坡国立大学） ； University of Chinese Academy of Sciences（中国科学院大学） ； University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校） 纠错

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

ResNetVLLM-2: Addressing ResNetVLLM's Multi-Modal Hallucinations

Ahmad Khalil, Mahmoud Khalil, Alioune Ngom

机构 * University of Windsor（温莎大学） 纠错

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

Amplify Initiative: Building A Localized Data Platform for Globalized AI

Qazi Mamunur Rashid, Erin van Liemt, Tiffany Shih, Amber Ebinama, Karla Barrios Ramos, Madhurima Maji, Aishwarya Verma, Charu Kalia, Jamila Smith-Loud, Joyce Nakatumba-Nabende, Rehema Baguma, Andrew Katumba, Chodrine Mutebi, Jagen Marvin, Eric Peter Wairagala, Mugizi Bruce, Peter Oketta, Lawrence Nderu, Obichi Obiajunwa, Abigail Oppong, Michael Zimba, Data Authors

机构 * Google Research USA（谷歌研究美国分公司） ； Makerere University（Makerere大学） ； Google Research India（谷歌研究印度） ； Jommo Kenyatta University Agriculture & Technology（乔姆摩肯雅塔大学农业与技术） ； Hutzpa Innovations Nigeria（Hutzpa创新尼日利亚） ； Malawi University of Science & Technology（马拉维科学技术大学） 纠错

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

Real-World Depth Recovery via Structure Uncertainty Modeling and Inaccurate GT Depth Fitting

Delong Suzhang, Meng Yang

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

Hallucination-Aware Generative Pretrained Transformer for Cooperative Aerial Mobility Control

Hyojun Ahn, Seungcheol Oh, Gyu Seon Kim, Soyi Jung, Soohyun Park, Joongheon Kim

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

Capturing AI's Attention: Physics of Repetition, Hallucination, Bias and Beyond

Frank Yingjie Huo, Neil F. Johnson

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

Comments Comments welcome to neiljohnson@gwu.edu

On Training-Conditional Conformal Prediction and Binomial Proportion Confidence Intervals

Rudi Coppola, Manuel Mazo

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

Journal ref Transactions on Machine Learning Research (2025)

UAC: Uncertainty-Aware Calibration of Neural Networks for Gesture Detection

Farida Al Haddad, Yuxin Wang, Malcolm Mielle

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

Comments 12 pages, 2 figures

Medical Reasoning in LLMs: An In-Depth Analysis of DeepSeek R1

Birger Moell, Fredrik Sand Aronsson, Sanian Akbar

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

Latent Lexical Projection in Large Language Models: A Novel Approach to Implicit Representation Refinement

Ziad Shaker, Brendan Ashdown, Hugo Fitzalan, Alistair Heathcote, Jocasta Huntington

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

Comments arXiv admin note: This paper has been withdrawn by arXiv due to disputed and unverifiable authorship

GAN-enhanced Simulation-driven DNN Testing in Absence of Ground Truth

Mohammed Attaoui, Fabrizio Pastore

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

Comments 15 pages, 8 figures, 13 tables

Uncertainty-Aware Hybrid Inference with On-Device Small and Remote Large Language Models

Seungeun Oh, Jinhyuk Kim, Jihong Park, Seung-Woo Ko, Tony Q. S. Quek, Seong-Lyun Kim

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

Comments 7 pages, 6 figures; to be presented at IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN) 2025

Investigating Human-Aligned Large Language Model Uncertainty

Kyle Moore, Jesse Roberts, Daryl Watson, Pamela Wisniewski

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

DSVD: Dynamic Self-Verify Decoding for Faithful Generation in Large Language Models

YiQiu Guo, Yuchen Yang, Zhe Chen, Pingjie Wang, Yusheng Liao, Ya Zhang, Yanfeng Wang, Yu Wang

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

Can Structured Data Reduce Epistemic Uncertainty?

Shriram M S, Sushmitha S, Gayathri K S, Shahina A

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

Comments Presented at NeLaMKRR@KR, 2024 (arXiv:2410.05339)

Intent Tagging: Exploring Micro-Prompting Interactions for Supporting Granular Human-GenAI Co-Creation Workflows

Frederic Gmeiner, Nicolai Marquardt, Michael Bentley, Hugo Romat, Michel Pahud, David Brown, Asta Roseway, Nikolas Martelaro, Kenneth Holstein, Ken Hinckley, Nathalie Riche

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

Comments 31 pages, 30 figures, 3 tables. To appear in the Proceedings of the 2025 ACM CHI Conference on Human Factors in Computing Systems, Yokohama, Japan

Heterogeneous Decision Making in Mixed Traffic: Uncertainty-aware Planning and Bounded Rationality

Hang Wang, Qiaoyi Fang, Junshan Zhang

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

Comments CPAL 2025

Random-Set Neural Networks (RS-NN)

Shireen Kudukkil Manchingal, Muhammad Mubashar, Kaizheng Wang, Keivan Shariatmadar, Fabio Cuzzolin

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

Comments Published as a conference paper at the Thirteenth International Conference on Learning Representations (ICLR 2025)

Uncertainty Quantification and Decomposition for LLM-based Recommendation

Wonbin Kweon, Sanghwan Jang, SeongKu Kang, Hwanjo Yu

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

Comments WWW 2025

Advancing Geological Carbon Storage Monitoring With 3d Digital Shadow Technology

Abhinav Prakash Gahlot, Rafael Orozco, Felix J. Herrmann

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

From Uncertainty to Trust: Kernel Dropout for AI-Powered Medical Predictions

Ubaid Azam, Imran Razzak, Shelly Vishwakarma, Hakim Hacid, Dell Zhang, Shoaib Jameel

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

Trustworthiness in Stochastic Systems: Towards Opening the Black Box

Jennifer Chien, David Danks

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

Comments 22 pages, 2 figures

Approximate Message Passing for Bayesian Neural Networks

Romeo Sommerfeld, Christian Helms, Ralf Herbrich

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

Comments for code see https://github.com/christian-helms/mpbnns.git

Improving Contextual Faithfulness of Large Language Models via Retrieval Heads-Induced Optimization

Lei Huang, Xiaocheng Feng, Weitao Ma, Yuchun Fan, Xiachong Feng, Yangfan Ye, Weihong Zhong, Yuxuan Gu, Baoxin Wang, Dayong Wu, Guoping Hu, Bing Qin

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

Comments Submitted to ARR October 2024

Uncertainty Guarantees on Automated Precision Weeding using Conformal Prediction

Paul Melki, Lionel Bombrun, Boubacar Diallo, Jérôme Dias, Jean-Pierre da Costa

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