TraceDet: Hallucination Detection from the Decoding Trace of Diffusion Large Language Models
Department of Engineering Science, University of Oxford, UK(牛津大学工程科学系);Hasso Plattner Institute, University of Potsdam(波茨坦大学哈索普兰特纳研究所);Carnegie Mellon University(卡内基梅隆大学);Mohamed bin Zayed University of Artificial Intelligence(姆罕默德·本·拉希德智能大学)
Diffusion large language models (D-LLMs) have recently emerged as a promising alternative to auto-regressive LLMs (AR-LLMs). However, the hallucination problem in D-LLMs remains underexplored, limiting their reliability in real-world applications. Existing hallucination detection methods are designed for AR-LLMs and rely on signals from single-step generation, making them ill-suited for D-LLMs where hallucination signals often emerge throughout the multi-step denoising process. To bridge this gap, we propose TraceDet, a novel framework that explicitly leverages the intermediate denoising steps of D-LLMs for hallucination detection. TraceDet models the denoising process as an action trace, with each action defined as the model's prediction over the cleaned response, conditioned on the previous intermediate output. By identifying the sub-trace that is maximally informative to the hallucinated responses, TraceDet leverages the key hallucination signals in the multi-step denoising process of D-LLMs for hallucination detection. Extensive experiments on various open source D-LLMs demonstrate that TraceDet consistently improves hallucination detection, achieving an average gain in AUROC of 15.2% compared to baselines.
Lower Layers Matter: Alleviating Hallucination via Multi-Layer Fusion Contrastive Decoding with Truthfulness Refocused
Sun Yat-Sen University(中山大学);Shenzhen MSU-BIT University(深圳MSU-BIT大学);UNSW, Sydney, NSW 2052, Australia(新南威尔士大学,悉尼,新南威尔士州,2052,澳大利亚);Shenzhen Key Laboratory for High Performance Data Mining, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences(深圳高性能数据挖掘关键实验室,深圳先进技术研究院,中国科学院)
Mitigating Hallucinations of Large Language Models in Medical Information Extraction via Contrastive Decoding
University of Science and Technology of China(中国科学技术大学);Jarvis Research Center, Tencent YouTu Lab(腾讯优图实验室贾维斯研究中心);Peking University(北京大学);Westlake University(西湖大学);City University of Hong Kong(香港城市大学)
DHI: Leveraging Diverse Hallucination Induction for Enhanced Contrastive Factuality Control in Large Language Models
School of Computer Science, Wuhan University(武汉大学计算机学院);Tsinghua Shenzhen International Graduate School, Tsinghua University(清华大学深圳国际研究生院);School of Artificial Intelligence, Wuhan University(武汉大学人工智能学院)
DeCoRe: Decoding by Contrasting Retrieval Heads to Mitigate Hallucinations
University of Edinburgh(爱丁堡大学);Centre for AI, Data Science & Artificial Intelligence, R&D, AstraZeneca(阿斯利康研发中心人工智能、数据科学与人工智能中心);University College London(伦敦大学学院)
In-Context Sharpness as Alerts: An Inner Representation Perspective for Hallucination Mitigation
City University of Hong Kong(香港城市大学);National University of Singapore(新加坡国立大学);Shanghai Jiao Tong University(上海交通大学);Stanford University(斯坦福大学);Penn State University(宾夕法尼亚州立大学);Hong Kong University of Science and Technology(香港科技大学)
DOPRA: Decoding Over-accumulation Penalization and Re-allocation in Specific Weighting Layer
Northeastern University(东北大学);Sydney Smart Technology College(悉尼智能科技学院);Shanghai Jiao Tong University(上海交通大学);School of Electronic Information and Electrical Engineering(电子信息与电气工程学院)
Exploring and Mitigating Fawning Hallucinations in Large Language Models
School of Medical Technology, Beijing Institute of Technology(北京理工大学医学技术学院);Artificial Intelligence Research Institute, Shenzhen MSU-BIT University(深圳MSU-BIT大学人工智能研究院);Tianjin University(天津大学);The Hong Kong University of Science(香港科学大学)
MMD-Flagger: Leveraging Maximum Mean Discrepancy to Detect Hallucinations
Université Côte d’Azur, CNRS, LJAD, France(法国埃克塞特大学、法国国家科学研究中心、LJAD研究所);Center for Artificial Intelligence and Data Science (CAIDAS)(人工智能与数据科学中心);Julius-Maximilans-Universität Würzburg, Germany(德国维尔茨堡约利乌斯-马克斯米利安大学)
Mitigating Object Hallucination via Robust Local Perception Search
Shanghai Artificial Intelligence Laboratory(上海人工智能实验室);Center for Future Media & School of Computer Science and Engineering, University of Electronic Science and Technology of China(未来媒体中心及电子科技大学计算机科学与工程学院)
Mitigating Video Hallucinations via Attention-Guided Representation Perturbation
School of Computer Science and Engineering, Beihang University(北京航空航天大学计算机科学与工程学院);Beijing Institute for General Artificial Intelligence, BIGAI(北京通用人工智能研究院);Sino-French Engineer School, Beihang University(北京航空航天大学中法工程师学院)
Mitigating Hallucinations in Large Vision-Language Models with Internal Fact-based Contrastive Decoding
School of Future Technology, Shanghai University(上海大学未来技术学院);Institute of Artificial Intelligence, Shanghai University(上海大学人工智能研究院);School of Mechatronic Engineering and Automation(机电工程与自动化学院)
Mixture of Decoding: An Attention-Inspired Adaptive Decoding Strategy to Mitigate Hallucinations in Large Vision-Language Models
New Laboratory of Pattern Recognition (NLPR), Institute of Automation, Chinese Academy of Sciences (CASIA)(模式识别新技术实验室,自动化研究所,中国科学院);School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院);Kuaishou Technology(快手科技);Nanjing University(南京大学)
本文提出 Mixture of Decoding(MoD),通过比较原始与被注意图像 token 的输出一致性,自适应采用互补或对比解码策略,从而在多个基准上有效缓解 LVLMs 幻觉。