大模型对齐与安全 - arXivDaily 专题

2505.23968 2025-06-02 cs.CR cs.AI cs.CY cs.LG stat.ML 67%

Confidential Guardian: Cryptographically Prohibiting the Abuse of Model Abstention

Stephan Rabanser, Ali Shahin Shamsabadi, Olive Franzese, Xiao Wang, Adrian Weller, Nicolas Papernot

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

Comments Proceedings of the 42nd International Conference on Machine Learning

2505.23854 2025-06-02 cs.CL cs.AI cs.LG 67%

Revisiting Uncertainty Estimation and Calibration of Large Language Models

Linwei Tao, Yi-Fan Yeh, Minjing Dong, Tao Huang, Philip Torr, Chang Xu

机构 * School of Computer Science University of Sydney（悉尼大学计算机科学学院）； City University of Hong Kong（香港城市大学）； Shanghai Jiao Tong University（上海交通大学）； Department of Engineering Science University of Oxford（牛津大学工程科学系）

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

2505.22655 2025-05-29 cs.LG cs.AI cs.CL 67%

Position: Uncertainty Quantification Needs Reassessment for Large-language Model Agents

Michael Kirchhof, Gjergji Kasneci, Enkelejda Kasneci

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

Comments Accepted at ICML 2025

2504.17671 2025-05-16 cs.CL cs.AI cs.LG 67%

Data-Driven Calibration of Prediction Sets in Large Vision-Language Models Based on Inductive Conformal Prediction

Yuanchang Ye, Weiyan Wen

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

Comments Accepted by ICIPCA 2025

2402.06529 2025-02-12 cs.AI cs.CL cs.LG 67%

Introspective Planning: Aligning Robots' Uncertainty with Inherent Task Ambiguity

Kaiqu Liang, Zixu Zhang, Jaime Fernández Fisac

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

Comments NeurIPS 2024

2502.04428 2025-02-10 cs.CL cs.AI cs.LG 67%

Confident or Seek Stronger: Exploring Uncertainty-Based On-device LLM Routing From Benchmarking to Generalization

Yu-Neng Chuang, Leisheng Yu, Guanchu Wang, Lizhe Zhang, Zirui Liu, Xuanting Cai, Yang Sui, Vladimir Braverman, Xia Hu

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

2308.12215 2025-01-28 cs.LG cs.CL cs.CY 67%

SoK: Machine Learning for Misinformation Detection

Madelyne Xiao, Jonathan Mayer

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

2412.07961 2024-12-12 cs.CL cs.AI cs.LG 67%

Forking Paths in Neural Text Generation

Eric Bigelow, Ari Holtzman, Hidenori Tanaka, Tomer Ullman

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

2412.00760 2024-12-03 eess.AS cs.AI cs.CL cs.ET cs.LG 67%

Automating Feedback Analysis in Surgical Training: Detection, Categorization, and Assessment

Firdavs Nasriddinov, Rafal Kocielnik, Arushi Gupta, Cherine Yang, Elyssa Wong, Anima Anandkumar, Andrew Hung

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

Comments Accepted as a proceedings paper at Machine Learning for Health 2024

2410.23054 2024-11-25 cs.LG cs.AI cs.CL cs.CV 67%

Controlling Language and Diffusion Models by Transporting Activations

Pau Rodriguez, Arno Blaas, Michal Klein, Luca Zappella, Nicholas Apostoloff, Marco Cuturi, Xavier Suau

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

2407.00499 2024-11-19 cs.CL cs.AI cs.LG 67%

ConU: Conformal Uncertainty in Large Language Models with Correctness Coverage Guarantees

Zhiyuan Wang, Jinhao Duan, Lu Cheng, Yue Zhang, Qingni Wang, Xiaoshuang Shi, Kaidi Xu, Hengtao Shen, Xiaofeng Zhu

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

Comments Accepted by EMNLP 2024 Findings

2402.14259 2024-11-19 cs.CL cs.AI cs.LG 67%

Word-Sequence Entropy: Towards Uncertainty Estimation in Free-Form Medical Question Answering Applications and Beyond

Zhiyuan Wang, Jinhao Duan, Chenxi Yuan, Qingyu Chen, Tianlong Chen, Yue Zhang, Ren Wang, Xiaoshuang Shi, Kaidi Xu

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

Comments Accepted by Engineering Applications of Artificial Intelligence

2406.04331 2024-11-06 cs.CL cs.AI cs.IR cs.LG 67%

PaCE: Parsimonious Concept Engineering for Large Language Models

Jinqi Luo, Tianjiao Ding, Kwan Ho Ryan Chan, Darshan Thaker, Aditya Chattopadhyay, Chris Callison-Burch, René Vidal

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

Comments Accepted in NeurIPS 2024. GitHub repository at https://github.com/peterljq/Parsimonious-Concept-Engineering

2407.02062 2024-10-28 cs.CL cs.AI cs.LG 67%

Are Data Augmentation Methods in Named Entity Recognition Applicable for Uncertainty Estimation?

Wataru Hashimoto, Hidetaka Kamigaito, Taro Watanabe

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

Comments Accepted to EMNLP 2024 main conference

2410.16592 2024-10-23 cs.LG cs.CL cs.CY 67%

ViMGuard: A Novel Multi-Modal System for Video Misinformation Guarding

Andrew Kan, Christopher Kan, Zaid Nabulsi

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

Comments 7 pages, 2 figures

2408.10474 2024-08-21 cs.SE cs.AI cs.CL cs.CR cs.LG 67%

LeCov: Multi-level Testing Criteria for Large Language Models

Xuan Xie, Jiayang Song, Yuheng Huang, Da Song, Fuyuan Zhang, Felix Juefei-Xu, Lei Ma

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

2408.07453 2024-08-15 cs.CL cs.AI cs.LG 67%

Fact or Fiction? Improving Fact Verification with Knowledge Graphs through Simplified Subgraph Retrievals

Tobias A. Opsahl

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

Comments 10 pages, 3 figures, appendix

2407.12836 2024-07-19 cs.CL cs.AI cs.LG 67%

OSPC: Artificial VLM Features for Hateful Meme Detection

Peter Grönquist

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

2406.00030 2024-06-04 cs.CL cs.AI cs.LG 67%

Large Language Model Pruning

Hanjuan Huang, Hao-Jia Song, Hsing-Kuo Pao

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

Comments 17 pages, 7 figures, 2 tables

2405.20003 2024-05-31 cs.LG cs.AI cs.CL 67%

Kernel Language Entropy: Fine-grained Uncertainty Quantification for LLMs from Semantic Similarities

Alexander Nikitin, Jannik Kossen, Yarin Gal, Pekka Marttinen

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

2309.13788 2024-04-25 cs.CL cs.AI cs.CR cs.HC cs.LG 67%

Can LLM-Generated Misinformation Be Detected?

Canyu Chen, Kai Shu

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

Comments Accepted to Proceedings of ICLR 2024. 9 pages for main paper, 40 pages including appendix. The code, results, dataset for this paper and more resources on "LLMs Meet Misinformation" have been released on the project website: https://llm-misinformation.github.io/

2404.08705 2024-04-16 cs.CL cs.AI cs.LG 67%

Introducing L2M3, A Multilingual Medical Large Language Model to Advance Health Equity in Low-Resource Regions

Agasthya Gangavarapu

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

2403.05973 2024-03-12 cs.CL cs.AI cs.LG 67%

Calibrating Large Language Models Using Their Generations Only

Dennis Ulmer, Martin Gubri, Hwaran Lee, Sangdoo Yun, Seong Joon Oh

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

2403.03170 2024-03-12 cs.MM cs.AI cs.CL cs.CV cs.CY 67%

SNIFFER: Multimodal Large Language Model for Explainable Out-of-Context Misinformation Detection

Peng Qi, Zehong Yan, Wynne Hsu, Mong Li Lee

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

Comments To appear in CVPR 2024

2311.09114 2024-02-27 cs.CL cs.AI cs.LG 67%

Ever: Mitigating Hallucination in Large Language Models through Real-Time Verification and Rectification

Haoqiang Kang, Juntong Ni, Huaxiu Yao

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

2311.09731 2024-02-19 cs.CL cs.AI cs.LG 67%

Examining LLMs' Uncertainty Expression Towards Questions Outside Parametric Knowledge

Genglin Liu, Xingyao Wang, Lifan Yuan, Yangyi Chen, Hao Peng

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

Comments 20 pages

2402.07023 2024-02-13 cs.CL cs.AI cs.CV cs.HC cs.LG 67%

Gemini Goes to Med School: Exploring the Capabilities of Multimodal Large Language Models on Medical Challenge Problems & Hallucinations

Ankit Pal, Malaikannan Sankarasubbu

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

Comments Preprint version, Under Review

2305.04782 2023-12-05 cs.CL cs.AI cs.LG 67%

HistAlign: Improving Context Dependency in Language Generation by Aligning with History

David Wan, Shiyue Zhang, Mohit Bansal

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

Comments EMNLP 2023 (20 pages)

2311.08401 2023-11-15 cs.CL cs.AI cs.LG 67%

Fine-tuning Language Models for Factuality

Katherine Tian, Eric Mitchell, Huaxiu Yao, Christopher D. Manning, Chelsea Finn

专题命中幻觉与事实性：RLHF(abstract)；分类 cs.CL、cs.AI、cs.LG

2309.10346 2023-09-20 cs.LG cs.AI cs.CL 67%

Explaining Agent Behavior with Large Language Models

Xijia Zhang, Yue Guo, Simon Stepputtis, Katia Sycara, Joseph Campbell

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

Comments Human Multi-Robot Interaction Workshop at IROS 2023

1. 幻觉与事实性 1732 篇

Confidential Guardian: Cryptographically Prohibiting the Abuse of Model Abstention

Revisiting Uncertainty Estimation and Calibration of Large Language Models

Position: Uncertainty Quantification Needs Reassessment for Large-language Model Agents

Data-Driven Calibration of Prediction Sets in Large Vision-Language Models Based on Inductive Conformal Prediction

Introspective Planning: Aligning Robots' Uncertainty with Inherent Task Ambiguity

Confident or Seek Stronger: Exploring Uncertainty-Based On-device LLM Routing From Benchmarking to Generalization

SoK: Machine Learning for Misinformation Detection

Forking Paths in Neural Text Generation

Automating Feedback Analysis in Surgical Training: Detection, Categorization, and Assessment

Controlling Language and Diffusion Models by Transporting Activations

ConU: Conformal Uncertainty in Large Language Models with Correctness Coverage Guarantees

Word-Sequence Entropy: Towards Uncertainty Estimation in Free-Form Medical Question Answering Applications and Beyond

PaCE: Parsimonious Concept Engineering for Large Language Models

Are Data Augmentation Methods in Named Entity Recognition Applicable for Uncertainty Estimation?

ViMGuard: A Novel Multi-Modal System for Video Misinformation Guarding

LeCov: Multi-level Testing Criteria for Large Language Models

Fact or Fiction? Improving Fact Verification with Knowledge Graphs through Simplified Subgraph Retrievals

OSPC: Artificial VLM Features for Hateful Meme Detection

Large Language Model Pruning

Kernel Language Entropy: Fine-grained Uncertainty Quantification for LLMs from Semantic Similarities

Can LLM-Generated Misinformation Be Detected?

Introducing L2M3, A Multilingual Medical Large Language Model to Advance Health Equity in Low-Resource Regions

Calibrating Large Language Models Using Their Generations Only

SNIFFER: Multimodal Large Language Model for Explainable Out-of-Context Misinformation Detection

Ever: Mitigating Hallucination in Large Language Models through Real-Time Verification and Rectification

Examining LLMs' Uncertainty Expression Towards Questions Outside Parametric Knowledge

Gemini Goes to Med School: Exploring the Capabilities of Multimodal Large Language Models on Medical Challenge Problems & Hallucinations

HistAlign: Improving Context Dependency in Language Generation by Aligning with History

Fine-tuning Language Models for Factuality

Explaining Agent Behavior with Large Language Models