HIVE: Hidden-Evidence Verification for Hallucination Detection in Diffusion Large Language Models
Not on the map. Why?
Abstract (arXiv)
Diffusion large language models generate text through iterative denoising, exposing hidden trajectories that may contain reliability signals beyond the final output. We propose HIVE, which compresses trajectory hidden states, selects informative step-layer evidence, and conditions a verifier through continuous prefix embeddings to produce a hallucination score and structured diagnostics. Across two D-LLMs and three QA benchmarks, HIVE outperforms eight established baselines and a verifier-backbone-matched text-only control in all six settings. Relative to text-only verification, hidden-evidence conditioning improves AUROC by 1.73--4.60 points and AUPRC by 1.10--3.62 points, with average gains of 3.15 and 2.28 points, respectively. Ablations, evidence interventions, and cross-dataset transfer further support the complementary value of fine-grained hidden trajectory evidence.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 44 of its 57 references that are in the map. First posted on arXiv 28 Apr 2026.
Builds on
44 of 57 references.
References · 44
- 2024Qwen2.5 Technical Reportcited by 4,942
- 2021Prefix-Tuning: Optimizing Continuous Prompts for Generationcited by 3,915
- 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
- 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,955
- 2022Language Models (Mostly) Know What They Knowcited by 1,856
- 2023FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generationcited by 1,282
- 2019BERT Rediscovers the Classical NLP Pipelinecited by 1,059
- 2022Diffusion-LM Improves Controllable Text Generationcited by 1,027
- 2025Large Language Diffusion Modelscited by 912
- 2023Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generationcited by 819
- 2023SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Modelscited by 810
- 2022Discovering Latent Knowledge in Language Models Without Supervisioncited by 767
32 more
- 2023Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Modelscited by 698
- 2023The Internal State of an LLM Knows When It's Lyingcited by 659
- 2025Dream 7B: Diffusion Large Language Modelscited by 544
- 2022DiffuSeq: Sequence to Sequence Text Generation with Diffusion Modelscited by 450
- 2024INSIDE: LLMs' Internal States Retain the Power of Hallucination Detectioncited by 323
- 2023Generating with Confidence: Uncertainty Quantification for Black-box Large Language Modelscited by 293
- 2024Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMscited by 259
- 2024Lookback Lens: Detecting and Mitigating Contextual Hallucinations in Large Language Models Using Only Attention Mapscited by 142
- 2023LM-Polygraph: Uncertainty Estimation for Language Modelscited by 131
- 2022Latent Diffusion for Language Generationcited by 115
- 2024HaloScope: Harnessing Unlabeled LLM Generations for Hallucination Detectioncited by 99
- 2023SAC3: Reliable Hallucination Detection in Black-Box Language Models via Semantic-aware Cross-check Consistencycited by 90
- 2024Unsupervised Real-Time Hallucination Detection based on the Internal States of Large Language Modelscited by 86
- 2023Knowing What LLMs DO NOT Know: A Simple Yet Effective Self-Detection Methodcited by 51
- 2025Steer LLM Latents for Hallucination Detectioncited by 50
- 2024On Large Language Models' Hallucination with Regard to Known Factscited by 43
- 2025Scaling Behavior of Discrete Diffusion Language Modelscited by 35
- 2025ICR Probe: Tracking Hidden State Dynamics for Reliable Hallucination Detection in LLMscited by 32
- 2025Uncertainty Quantification for LLMs through Minimum Bayes Risk: Bridging Confidence and Consistencycited by 30
- 2024Diffusion Guided Language Modelingcited by 15
- 2026DynHD: Hallucination Detection for Diffusion Large Language Models via Denoising Dynamics Deviation Learningcited by 15
- 2025FactSelfCheck: Fact-Level Black-Box Hallucination Detection for LLMscited by 14
- 2023OPORP: One Permutation + One Random Projectioncited by 12
- 2024Unconditional Truthfulness: Learning Unconditional Uncertainty of Large Language Modelscited by 9
- 2024Prompt-Guided Internal States for Hallucination Detection of Large Language Modelscited by 8
- 2025TraceDet: Hallucination Detection from the Decoding Trace of Diffusion Large Language Modelscited by 8
- 2026HALT: Hallucination Assessment via Log-probs as Time seriescited by 8
- 2026TDGNet: Hallucination Detection in Diffusion Language Models via Temporal Dynamic Graphscited by 8
- 2026Diffusion Language Models Are Natively Length-Awarecited by 5
- 2025Consistency Is the Key: Detecting Hallucinations in LLM Generated Text By Checking Inconsistencies About Key Factscited by 4
- 2026Early Decisions Matter: Proximity Bias and Initial Trajectory Shaping in Non-Autoregressive Diffusion Language Modelscited by 4
- 2026Lost in Diffusion: Uncovering Hallucination Patterns and Failure Modes in Diffusion Large Language Modelscited by 2
Built on by
Citing papers per year: 2026: 1.
1 paper.