mlsanityData to 14 Oct 2026

HalluScope: Fine-grained Hallucination Diagnosis for Multimodal Large Language Models

arXiv 2607.21105 ↗First posted 23 Jul 2026cs.CVcited by 0

On the map: Multimodal LLMs / vision-language models → Hallucination mitigation in LVLMs

Abstract (arXiv)

Although Multimodal Large Language Models have achieved strong performance across a wide range of vision-language tasks, they still suffer from hallucinations, where model outputs become inconsistent with the visual content, textual context, or commonsense knowledge. Existing studies primarily address this problem through coarse-grained detection. However, these approaches often provide insufficient diagnostic information for understanding hallucination types and supporting downstream hallucination mitigation. To bridge this gap, we propose fine-grained hallucination diagnosis for MLLMs, a new unified task that jointly performs hallucination detection, classification, and interpretable explanation generation. We develop an automated data generation pipeline and construct HalluScope-30K, a large-scale diagnostic dataset covering eight sources and five task categories. Based on this dataset, we design a multi-granular joint reward function and train two diagnosis models, HalluScope-4B and HalluScope-8B, which achieve state-of-the-art performance on both the MHALO benchmark and our fine-grained hallucination classification benchmark. Notably, detection and classification are mutually beneficial under joint optimization. Furthermore, diagnosis-driven feedback experiments show that the fine-grained diagnostic explanations produced by our model effectively guide target models to correct their hallucinations, with full diagnosis substantially outperforming all baselines on both Qwen3-VL-8B-Instruct and LLaVA-1.5-7B. Our code, data, and models are available at https://github.com/wkinglin/HalluScope.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Hallucination mitigation in LVLMs, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 28 of its 59 references that are in the map. First posted on arXiv 23 Jul 2026.

Builds on

28 of 59 references.

Closest ancestors · 14

Same direction, newest first

  1. 2025A Survey of Multimodal Hallucination Evaluation and Detection25 Jul
  2. 2025MTRE: Multi-Token Reliability Estimation for Hallucination Detection in VLMs16 May
  3. 2024Nullu: Mitigating Object Hallucinations in Large Vision-Language Models via HalluSpace Projection18 Dec
  4. 2024VL-Uncertainty: Detecting Hallucination in Large Vision-Language Model via Uncertainty Estimation18 Nov
  5. 2024Pre-Training Multimodal Hallucination Detectors with Corrupted Grounding Data30 Aug
  6. 2024Pelican: Correcting Hallucination in Vision-LLMs via Claim Decomposition and Program of Thought Verification2 Jul
  7. 2024MetaToken: Detecting Hallucination in Image Descriptions by Meta Classification29 May
  8. 2024Detecting and Mitigating Hallucination in Large Vision Language Models via Fine-Grained AI Feedback22 Apr
  9. 2024Mitigating Hallucinations in Large Vision-Language Models with Instruction Contrastive Decoding27 Mar
  10. 2024IBD: Alleviating Hallucinations in Large Vision-Language Models via Image-Biased Decoding28 Feb
  11. 2023RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedback1 Dec
  12. 2023Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decoding28 Nov
2 more
  1. 2023Beyond Hallucinations: Enhancing LVLMs through Hallucination-Aware Direct Preference Optimization28 Nov
  2. 2023AMBER: An LLM-free Multi-dimensional Benchmark for MLLMs Hallucination Evaluation13 Nov

Other references · 14

  1. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,442
  2. 2015VQA: Visual Question Answeringcited by 2,506
  3. 2023MMBench: Is Your Multi-modal Model an All-around Player?cited by 2,141
  4. 2015Microsoft COCO Captions: Data Collection and Evaluation Servercited by 1,462
  5. 2023LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Daycited by 1,305
  6. 2024ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Toolscited by 1,121
  7. 2023Kosmos-2: Grounding Multimodal Large Language Models to the Worldcited by 1,089
  8. 2023CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquingcited by 783
  9. 2023Aligning Large Multimodal Models with Factually Augmented RLHFcited by 712
  10. 2024Fine-grained Hallucination Detection and Editing for Language Modelscited by 149
  11. 2024Exploring the Reasoning Abilities of Multimodal Large Language Models (MLLMs): A Comprehensive Survey on Emerging Trends in Multimodal Reasoningcited by 136
  12. 2024A Comprehensive Survey of Hallucination in Large Language, Image, Video and Audio Foundation Modelscited by 100
2 more
  1. 2025OpenMMReasoner: Pushing the Frontiers for Multimodal Reasoning with an Open and General Recipecited by 37
  2. 2024Target Span Detection for Implicit Harmful Contentcited by 8

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Scalable Token-Level Hallucination Detection in Large Language Models
  2. 2025From Detection to Diagnosis: Advancing Hallucination Analysis with Automated Data Synthesis
  3. 2024Unified Hallucination Detection for Multimodal Large Language Models
  4. 2026ReactBench: A Cause-Driven Benchmark for Multimodal Hallucination via Systematic Evaluation
  5. 2026Beyond the Global Scores: Fine-Grained Token Grounding as a Robust Detector of LVLM Hallucinations
  6. 2026OmniHallu: Unified Hallucination Detection for Cross-Modal Comprehension and Generation in Multimodal Large Language Models
  7. 2026RAGognizer: Hallucination-Aware Fine-Tuning via Detection Head Integration
  8. 2025ZINA: Multimodal Fine-grained Hallucination Detection and Editing