大模型对齐与安全 - arXivDaily 专题

2503.06472 2026-01-07 cs.CV cs.MM 50%

CalliReader: Contextualizing Chinese Calligraphy via an Embedding-Aligned Vision-Language Model

CalliReader：通过嵌入对齐的视觉-语言模型 contextualizing 中文书法

Yuxuan Luo, Jiaqi Tang, Chenyi Huang, Feiyang Hao, Zhouhui Lian

机构 * Wangxuan Institute of Computer Technology, Peking University（北京大学计算机技术研究院）

专题命中幻觉与事实性：alignment(abstract)

AI总结 CalliReader通过字符级切片、视觉-文本对齐和嵌入指令微调，解决中文书法的上下文识别问题，实现比现有方法和专业书法家更高的准确性与更低的幻觉率。

Comments 11 pages

Journal ref Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) 2025

2512.18671 2025-12-23 cs.CV 50%

SmartSight: Mitigating Hallucination in Video-LLMs Without Compromising Video Understanding via Temporal Attention Collapse

SmartSight: 通过时间注意力崩溃缓解视频大语言模型中的幻觉而不影响视频理解

Yiming Sun, Mi Zhang, Feifei Li, Geng Hong, Min Yang

专题命中幻觉与事实性：safety(abstract)

AI总结 SmartSight通过时间注意力崩溃技术，在不牺牲视频理解能力的前提下，有效降低视频大语言模型的幻觉问题。

Comments AAAI26 accepted

2508.17201 2025-12-19 physics.med-ph 50%

Root Cause Analysis of Radiation Oncology Incidents Using Large Language Models

利用大语言模型进行放射肿瘤事件的根本原因分析

Yuntao Wang, Mariluz De Ornelas, Matthew T. Studenski, Elizabeth Bossart, Siamak P. Najad-Davarani, Yunze Yang

专题命中幻觉与事实性：safety(abstract)

AI总结利用大语言模型分析放射肿瘤事件的根本原因，评估其在提升患者安全中的应用潜力。

2512.11096 2025-12-15 cs.HC 50%

Your plan may succeed, but what about failure? Investigating how people use ChatGPT for long-term life task planning

你的计划可能成功，但失败呢？探究人们如何利用ChatGPT进行长期生活任务规划

Ben Wang, Jiqun Liu

专题命中幻觉与事实性：trustworthy(abstract)

AI总结研究探讨人们如何利用ChatGPT进行长期生活任务规划，发现AI在结构化目标和生成想法方面有帮助，但输出缺乏个性化和适应性，需用户主动调整。

2512.09249 2025-12-11 cond-mat.mtrl-sci 50%

Auto-3DPFM: Automating Polarization-Vector Mapping at the Nanoscale

Auto-3DPFM：在纳米尺度上自动化极化矢量映射

Ralph Bulanadi, Marti Checa, Michelle Wang, Franck Rothen, John Lasseter, Sumner B. Harris, Daniel Sando, Valanoor Nagarajan, Liam Collins, Stephen Jesse, Rama Vasudevan, Yongtao Liu

专题命中幻觉与事实性：alignment(abstract)

AI总结 Auto-3DPFM通过自动化技术实现纳米尺度极化矢量的高精度表征，提升铁电材料研究的效率与准确性。

2508.04679 2025-12-11 cs.HC 50%

MisVisFix: An Interactive Dashboard for Detecting, Explaining, and Correcting Misleading Visualizations using Large Language Models

MisVisFix: 一种利用大语言模型检测、解释和纠正误导性可视化信息的交互式仪表板

Amit Kumar Das, Klaus Mueller

专题命中幻觉与事实性：trustworthy(abstract)

AI总结 MisVisFix利用大语言模型提供交互式工具，用于检测、解释和纠正误导性可视化信息，提升数据解读的准确性和可信度。

Comments 11 pages, 6 figures. Accepted at IEEE VIS: Visualization & Visual Analytics 2025 conference, November 2-7, 2025, Vienna, Austria

Journal ref IEEE Transactions on Visualization and Computer Graphics (TVCG), PrePrints 5555, pp. 1-11, 2025

2511.22019 2025-12-09 cs.CV 50%

Intra-Class Probabilistic Embeddings for Uncertainty Estimation in Vision-Language Models

类内概率嵌入用于视觉-语言模型中的不确定性估计

Zhenxiang Lin, Maryam Haghighat, Will Browne, Dimity Miller

机构 * Queensland University of Technology（昆士兰理工大学）

专题命中幻觉与事实性：safety(abstract)

AI总结本研究提出一种无需训练的后处理方法，通过类内概率嵌入提升视觉-语言模型的不确定性估计，有效检测错误预测。

Comments Accepted at the IEEE/CVF Winter Conference on Applications of Computer Vision 2026

2510.22171 2025-12-02 cs.CV 50%

HARMONY: Hidden Activation Representations and Model Output-Aware Uncertainty Estimation for Vision-Language Models

HARMONY：隐藏的激活表示和模型输出感知的不确定性估计用于视觉-语言模型

Erum Mushtaq, Zalan Fabian, Yavuz Faruk Bakman, Anil Ramakrishna, Mahdi Soltanolkotabi, Salman Avestimehr

机构 * University of Southern California（南加州大学）； Amazon AGI（亚马逊人工智能实验室）

专题命中幻觉与事实性：alignment(abstract)

AI总结 HARMONY通过整合生成token、模型输出不确定性分数和隐藏表示，提升视觉-语言模型的不确定性估计性能。

2511.22746 2025-12-01 cs.HC 50%

Epistemic Fragility in Large Language Models: Prompt Framing Systematically Modulates Misinformation Correction

大型语言模型中的知识脆弱性：提示框架系统性地调节 misinformation 的纠正

Sekoul Krastev, Hilary Sweatman, Anni Sternisko, Steve Rathje

专题命中幻觉与事实性：alignment(abstract)

AI总结研究发现提示框架对大型语言模型纠正虚假信息的能力有显著影响，揭示了模型在知识准确性上的脆弱性。

2511.19806 2025-11-26 cs.CV 50%

Reading Between the Lines: Abstaining from VLM-Generated OCR Errors via Latent Representation Probes

在行之间阅读：通过潜在表示探测避免VLM生成的OCR错误

Jihan Yao, Achin Kulshrestha, Nathalie Rauschmayr, Reed Roberts, Banghua Zhu, Yulia Tsvetkov, Federico Tombari

机构 * University of Washington（华盛顿大学）； Google（谷歌）

专题命中幻觉与事实性：safety(abstract)

AI总结本文提出通过潜在表示探测提升VLM在OCR任务中的回避准确性，通过内部状态检测置信度信号，提高系统可靠性。

2511.13533 2025-11-25 cs.CV 50%

Minimax Multi-Target Conformal Prediction with Applications to Imaging Inverse Problems

最小最大多目标符合预测及其在成像反问题中的应用

Jeffrey Wen, Rizwan Ahmad, Philip Schniter

机构 * Department of Electrical and Computer Engineering（电气与计算机工程系）； The Ohio State University（俄亥俄州立大学）； Department of Biomedical Engineering（生物医学工程系）

专题命中幻觉与事实性：safety(abstract)

AI总结本文提出了一种渐近最小最大方法用于多目标符合预测，应用于多指标盲图像质量评估、多任务不确定性量化和多轮测量获取，通过合成和MRI数据验证了其有效性。

Journal ref Transactions on Machine Learning Research, 11/2025. https://openreview.net/forum?id=53FEYwDQK0

2511.17875 2025-11-25 math.OC 50%

Modeling and Calibration of Supplier Selection Problem in Freight Agent-Based Simulations

货运 agent 仿真实验中供应商选择问题的建模与校准

Abdelrahman Ismael, Taner Cokyasar

专题命中幻觉与事实性：alignment(abstract)

AI总结本研究提出了一种大规模校准的 agent 仿真实验模型，用于建模和校准供应商选择问题，以提高货运网络流的预测精度和供应链的韧性。

2511.11984 2025-11-18 cs.CV 50%

From Classification to Cross-Modal Understanding: Leveraging Vision-Language Models for Fine-Grained Renal Pathology

Zhenhao Guo, Rachit Saluja, Tianyuan Yao, Quan Liu, Junchao Zhu, Haibo Wang, Daniel Reisenbüchler, Yuankai Huo, Benjamin Liechty, David J. Pisapia, Kenji Ikemura, Steven Salvatoree, Surya Seshane, Mert R. Sabuncu, Yihe Yang, Ruining Deng

机构 * New York University（纽约大学）； Cornell Tech（康奈尔科技）； Vanderbilt University（范德比大学）； Carnegie Mellon University（卡内基梅隆大学）； University of Regensburg（莱茵河畔大学）； Weill Cornell Medicine（韦尔·科恩医学中心）； Northwell Health（北well健康）

专题命中幻觉与事实性：alignment(abstract)

2511.11106 2025-11-17 cs.MM cs.CV cs.SD 50%

AccKV: Towards Efficient Audio-Video LLMs Inference via Adaptive-Focusing and Cross-Calibration KV Cache Optimization

Zhonghua Jiang, Kui Chen, Kunxi Li, Keting Yin, Yiyun Zhou, Zhaode Wang, Chengfei Lv, Shengyu Zhang

专题命中幻觉与事实性：alignment(abstract)

2511.10974 2025-11-17 cs.CV 50%

Preserving Cross-Modal Consistency for CLIP-based Class-Incremental Learning

Haoran Chen, Houze Xu, Micah Goldblum, Daoguo Dong, Zuxuan Wu

机构 * Institute of Trustworthy Embodied AI（可信具身人工智能研究院）； Fudan University（复旦大学）； Shanghai Collaborative Innovation Center of Intelligent Visual Computing（上海智能视觉计算协同创新中心）； Columbia University（哥伦比亚大学）

专题命中幻觉与事实性：alignment(abstract)

2507.19110 2025-11-14 cs.CV 50%

LISA: A Layer-wise Integration and Suppression Approach for Hallucination Mitigation in Multimodal Large Language Models

Zhihui Guo, Xin Man, Hui Xu, Jie Shao, Zhiguo Jiang, Xianchao Zhang, Heng Tao Shen

专题命中幻觉与事实性：alignment(abstract)

2511.06649 2025-11-11 cs.CE 50%

Differentiable Semantic Meta-Learning Framework for Long-Tail Motion Forecasting in Autonomous Driving

Bin Rao, Chengyue Wang, Haicheng Liao, Qianfang Wang, Yanchen Guan, Jiaxun Zhang, Xingcheng Liu, Meixin Zhu, Kanye Ye Wang, Zhenning Li

专题命中幻觉与事实性：safety(abstract)

Comments Accepted at AAAI-26 (AAAI 2026), Main Track

2511.06475 2025-11-11 cs.CV 50%

NOAH: Benchmarking Narrative Prior driven Hallucination and Omission in Video Large Language Models

Kyuho Lee, Euntae Kim, Jinwoo Choi, Buru Chang

机构 * Korea University（韩国大学）； Kyung Hee University（庆熙大学）

专题命中幻觉与事实性：trustworthy(abstract)

Comments 18 pages, 9 figures. Preprint

2507.06645 2025-11-10 q-bio.NC 50%

Quantifying Uncertainty in Error Consistency: Towards Reliable Behavioral Comparison of Classifiers

Thomas Klein, Sascha Meyen, Wieland Brendel, Felix A. Wichmann, Kristof Meding

专题命中幻觉与事实性：alignment(abstract)

Comments 25 pages, 14 figures

2411.16503 2025-10-28 cs.CV 50%

Noise Diffusion for Enhancing Semantic Faithfulness in Text-to-Image Synthesis

Boming Miao, Chunxiao Li, Xiaoxiao Wang, Andi Zhang, Rui Sun, Zizhe Wang, Yao Zhu

机构 * Beijing Normal University（北京师范大学）； University of Chinese Academy of Sciences（中国科学院大学）； University of Manchester（曼彻斯特大学）； The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））； Tsinghua University（清华大学）

专题命中幻觉与事实性：alignment(abstract)

Comments Updated author formatting; no substantive changes

2509.07759 2025-10-28 cs.IR 50%

A Survey of Long-Document Retrieval in the PLM and LLM Era

Minghan Li, Miyang Luo, Tianrui Lv, Yishuai Zhang, Siqi Zhao, Ercong Nie, Guodong Zhou

专题命中幻觉与事实性：alignment(abstract)

Comments 32 pages, 6 figures

2504.13169 2025-10-21 cs.CV 50%

Generate, but Verify: Reducing Hallucination in Vision-Language Models with Retrospective Resampling

Tsung-Han Wu, Heekyung Lee, Jiaxin Ge, Joseph E. Gonzalez, Trevor Darrell, David M. Chan

机构 * UC Berkeley（加州大学伯克利分校）； POSTECH

专题命中幻觉与事实性：safety(abstract)

Comments Accepted to NeurIPS 2025; Project Page: https://reverse-vlm.github.io

2510.08658 2025-10-13 econ.GN q-fin.EC 50%

When Truth Does Not Take on Its Shoes: How Misinformation Spreads in Chatrooms

Shuige Liu

专题命中幻觉与事实性：alignment(abstract)

2510.06070 2025-10-08 cs.CV 50%

There is More to Attention: Statistical Filtering Enhances Explanations in Vision Transformers

Meghna P Ayyar, Jenny Benois-Pineau, Akka Zemmari

机构 * LaBRI, CNRS, Univ. Bordeaux, UMR 5800（LaBRI、CNRS、波尔多大学、UMR 5800）

专题命中幻觉与事实性：alignment(abstract)

2510.05374 2025-10-08 eess.SY cs.SY 50%

Digital Twins for Intelligent Intersections: A Literature Review

Alben Rome Bagabaldo, Jürgen Hackl

专题命中幻觉与事实性：safety(abstract)

Comments 29 pages, 2 figures, under review at Transportation Research Interdisciplinary Perspectives journal

2509.20148 2025-10-07 cs.CV 50%

Smaller is Better: Enhancing Transparency in Vehicle AI Systems via Pruning

Sanish Suwal, Shaurya Garg, Dipkamal Bhusal, Michael Clifford, Nidhi Rastogi

专题命中幻觉与事实性：safety(abstract)

Comments 17 pages

2509.00320 2025-10-03 cs.CV 50%

TrimTokenator: Towards Adaptive Visual Token Pruning for Large Multimodal Models

Hao Zhang, Mengsi Lyu, Chenrui He, Yulong Ao, Yonghua Lin

机构 * Beijing Academy of Artificial Intelligence (BAAI)（北京人工智能研究院）

专题命中幻觉与事实性：alignment(abstract)

Comments 15 pages

2510.01126 2025-10-02 cs.CV cs.RO 50%

Strategic Fusion of Vision Language Models: Shapley-Credited Context-Aware Dawid-Skene for Multi-Label Tasks in Autonomous Driving

Yuxiang Feng, Keyang Zhang, Hassane Ouchouid, Ashwil Kaniamparambil, Ioannis Souflas, Panagiotis Angeloudis

机构 * Centre for Transport Engineering and Modelling, Department of Civil and Environmental Engineering, Imperial College London（交通工程与建模中心，土木与环境工程系，帝国理工学院伦敦分校）； ELM Europe（ELM欧洲）

专题命中幻觉与事实性：safety(abstract)

Comments 8 pages

2510.00721 2025-10-02 physics.chem-ph 50%

Flexible Uncertainty Calibration for Machine-Learned Interatomic Potentials

Cheuk Hin Ho, Christoph Ortner, Yangshuai Wang

专题命中幻觉与事实性：trustworthy(abstract)

2505.07618 2025-09-30 cs.IR 50%

KAQG: A Knowledge-Graph-Enhanced RAG for Difficulty-Controlled Question Generation

Ching Han Chen, Ming Fang Shiu

专题命中幻觉与事实性：alignment(abstract)

Comments 10 pages, 4 figures and 6 tables

1. 幻觉与事实性 1756 篇

CalliReader: Contextualizing Chinese Calligraphy via an Embedding-Aligned Vision-Language Model

SmartSight: Mitigating Hallucination in Video-LLMs Without Compromising Video Understanding via Temporal Attention Collapse

Root Cause Analysis of Radiation Oncology Incidents Using Large Language Models

Your plan may succeed, but what about failure? Investigating how people use ChatGPT for long-term life task planning

Auto-3DPFM: Automating Polarization-Vector Mapping at the Nanoscale

MisVisFix: An Interactive Dashboard for Detecting, Explaining, and Correcting Misleading Visualizations using Large Language Models

Intra-Class Probabilistic Embeddings for Uncertainty Estimation in Vision-Language Models

HARMONY: Hidden Activation Representations and Model Output-Aware Uncertainty Estimation for Vision-Language Models

Epistemic Fragility in Large Language Models: Prompt Framing Systematically Modulates Misinformation Correction

Reading Between the Lines: Abstaining from VLM-Generated OCR Errors via Latent Representation Probes

Minimax Multi-Target Conformal Prediction with Applications to Imaging Inverse Problems

Modeling and Calibration of Supplier Selection Problem in Freight Agent-Based Simulations

From Classification to Cross-Modal Understanding: Leveraging Vision-Language Models for Fine-Grained Renal Pathology

AccKV: Towards Efficient Audio-Video LLMs Inference via Adaptive-Focusing and Cross-Calibration KV Cache Optimization

Preserving Cross-Modal Consistency for CLIP-based Class-Incremental Learning

LISA: A Layer-wise Integration and Suppression Approach for Hallucination Mitigation in Multimodal Large Language Models

Differentiable Semantic Meta-Learning Framework for Long-Tail Motion Forecasting in Autonomous Driving

NOAH: Benchmarking Narrative Prior driven Hallucination and Omission in Video Large Language Models

Quantifying Uncertainty in Error Consistency: Towards Reliable Behavioral Comparison of Classifiers

Noise Diffusion for Enhancing Semantic Faithfulness in Text-to-Image Synthesis

A Survey of Long-Document Retrieval in the PLM and LLM Era

Generate, but Verify: Reducing Hallucination in Vision-Language Models with Retrospective Resampling

When Truth Does Not Take on Its Shoes: How Misinformation Spreads in Chatrooms

There is More to Attention: Statistical Filtering Enhances Explanations in Vision Transformers

Digital Twins for Intelligent Intersections: A Literature Review

Smaller is Better: Enhancing Transparency in Vehicle AI Systems via Pruning

TrimTokenator: Towards Adaptive Visual Token Pruning for Large Multimodal Models

Strategic Fusion of Vision Language Models: Shapley-Credited Context-Aware Dawid-Skene for Multi-Label Tasks in Autonomous Driving

Flexible Uncertainty Calibration for Machine-Learned Interatomic Potentials

KAQG: A Knowledge-Graph-Enhanced RAG for Difficulty-Controlled Question Generation