mlsanityData to 14 Oct 2026

TraceDet: Hallucination Detection from the Decoding Trace of Diffusion Large Language Models

arXiv 2510.01274 ↗First posted 30 Sep 2025cs.CL cs.LGcited by 8

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Builds on 53 of its 60 references that are in the map. First posted on arXiv 30 Sep 2025.

Builds on

53 of 60 references.

References · 53

  1. 2017Attention Is All You Needcited by 60,361
  2. 2023GPT-4 Technical Reportcited by 18,396
  3. 2025Qwen3 Technical Reportcited by 8,820
  4. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  5. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
  6. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,955
  7. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,739
  8. 2018CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledgecited by 1,822
  9. 2023A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questionscited by 1,601
  10. 2023Inference-Time Intervention: Eliciting Truthful Answers from a Language Modelcited by 1,302
  11. 2023FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generationcited by 1,282
  12. 2022Diffusion Models: A Comprehensive Survey of Methods and Applicationscited by 1,114
38 more
  1. 2022Diffusion-LM Improves Controllable Text Generationcited by 1,027
  2. 2023Training Diffusion Models with Reinforcement Learningcited by 982
  3. 2023Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMscited by 974
  4. 2016Deep Variational Information Bottleneckcited by 930
  5. 2025Large Language Diffusion Modelscited by 912
  6. 2024Simple and Effective Masked Diffusion Language Modelscited by 848
  7. 2023Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generationcited by 819
  8. 2023SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Modelscited by 810
  9. 2015Deep Learning and the Information Bottleneck Principlecited by 802
  10. 2022Discovering Latent Knowledge in Language Models Without Supervisioncited by 767
  11. 2022Teaching Models to Express Their Uncertainty in Wordscited by 751
  12. 2023The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasetscited by 712
  13. 2023Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Modelscited by 698
  14. 2023The Internal State of an LLM Knows When It's Lyingcited by 659
  15. 2024Simplified and Generalized Masked Diffusion for Discrete Datacited by 626
  16. 2025Dream 7B: Diffusion Large Language Modelscited by 544
  17. 2023A Survey of Hallucination in Large Foundation Modelscited by 368
  18. 2024INSIDE: LLMs' Internal States Retain the Power of Hallucination Detectioncited by 323
  19. 2023Generating with Confidence: Uncertainty Quantification for Black-box Large Language Modelscited by 293
  20. 2024Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMscited by 259
  21. 2023FacTool: Factuality Detection in Generative AI -- A Tool Augmented Framework for Multi-Task and Multi-Domain Scenarioscited by 251
  22. 2025MMaDA: Multimodal Large Diffusion Language Modelscited by 250
  23. 2024LLMs Know More Than They Show: On the Intrinsic Representation of LLM Hallucinationscited by 244
  24. 2022Out-of-Distribution Detection and Selective Generation for Conditional Language Modelscited by 197
  25. 2025d1: Scaling Reasoning in Diffusion Large Language Models via Reinforcement Learningcited by 186
  26. 2024The pitfalls of next-token predictioncited by 160
  27. 2024ReDeEP: Detecting Hallucination in Retrieval-Augmented Generation via Mechanistic Interpretabilitycited by 101
  28. 2024HaloScope: Harnessing Unlabeled LLM Generations for Hallucination Detectioncited by 99
  29. 2019Significance-aware Information Bottleneck for Domain Adaptive Semantic Segmentationcited by 74
  30. 2025The Hidden Life of Tokens: Reducing Hallucination of Large Vision-Language Models via Visual Information Steeringcited by 66
  31. 2025Steer LLM Latents for Hallucination Detectioncited by 50
  32. 2023Truth Forest: Toward Multi-Scale Truthfulness in Large Language Models through Intervention without Tuningcited by 45
  33. 2023Encoding Time-Series Explanations through Self-Supervised Model Behavior Consistencycited by 42
  34. 2019BottleSum: Unsupervised and Self-supervised Sentence Summarization using the Information Bottleneck Principlecited by 39
  35. 2024TimeX++: Learning Time-Series Explanations with Information Bottleneckcited by 33
  36. 2024Procedural Content Generation in Games: A Survey with Insights on Emerging LLM Integrationcited by 22
  37. 2024An Information Bottleneck Perspective for Effective Noise Filtering on Retrieval-Augmented Generationcited by 16
  38. 2025Attention-guided Self-reflection for Zero-shot Hallucination Detection in Large Language Modelscited by 16

50 most cited of 53 All 53 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 8.

8 papers.

  1. 2026DynHD: Hallucination Detection for Diffusion Large Language Models via Denoising Dynamics Deviation Learningcited by 15
  2. 2026OSCAR: Orchestrated Self-verification and Cross-path Refinementcited by 2
  3. 2026HIVE: Hidden-Evidence Verification for Hallucination Detection in Diffusion Large Language Modelscited by 1
  4. 2026TRE: Training-Free Hallucination Detection for Diffusion Language Modelscited by 1
  5. 2026Uncertainty Quantification for Large Language Diffusion Modelscited by 0
  6. 2026Extracting Training Data from Diffusion Language Models via Infillingcited by 0
  7. 2026The Detectability Gap: Hidden Heterogeneity in Hallucination Detection Across Language Modelscited by 0
  8. 2026TRACESTEER: Trace Driven Activation Steering for Hallucination Mitigation in Diffusion Language Modelscited by 0