awesome-hallucination-mllm

KO | EN

gitlite — search

#python #hacktoberfest #react #ai #golang #android #deep-learning #nodejs #awesome-list #pytorch #security #ios #openai #open-source

awesome-hallucination-mllm

(★ 33)

MLLM hallucination, LVLM, LLM, Hallucination Mitigation, Training-free hallucination mitigation

awesome-hallucination-mllm Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Paper List

*Last updated: 2026-07-13 | Total papers: 289*

If you find this repository helpful for your research, we would appreciate citations to our related papers. We also warmly welcome discussions and suggestions: 1004392768@qq.com.

```bibtex
@article{lyu2026hallu_sae,
  title={Towards Interpretable Hallucination Analysis and Mitigation in LVLMs via Contrastive Neuron Steering},
  author={Lyu, Guangtao and Cheng, Xinyi and Liu, Qi and Xu, Chenghao and Yan, Jiexi and Yang, Muli and Fang, Fen and Deng, Cheng},
  journal={arXiv preprint arXiv:2602.00621},
  year={2026}
}
@article{lyu2025hallu_vdc,
  title={Revealing Perception and Generation Dynamics in LVLMs: Mitigating Hallucinations via Validated Dominance Correction},
  author={Lyu, Guangtao and Cheng, Xinyi and Xu, Chenghao and Liu, Qi and Yang, Muli and Fang, Fen and Chen, Huilin and Yan, Jiexi and Yang, Xu and Deng, Cheng},
  journal={arXiv preprint arXiv:2512.18813},
  year={2025}
}
@article{lyu2026hallu_pade,
  title={Revealing and Enhancing Core Visual Regions: Harnessing Internal Attention Dynamics for Hallucination Mitigation in LVLMs},
  author={Lyu, Guangtao and Liu, Qi and Xu, Chenghao and Yan, Jiexi and Yang, Muli and Li, Xueting and Fang, Fen and Deng, Cheng},
  journal={ACL Findings},
  year={2026}
}
```

## 2026
1. **Where Does Vision Meet Language? Understanding and Refining Visual Fusion in MLLMs via Contrastive Attention**  ,   Song et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2601.08151-b31b1b.svg)](https://arxiv.org/abs/2601.08151)

2. **When Prompts Override Vision: Prompt-Induced Hallucinations in LVLMs**  ,   Khayatan et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.21911-b31b1b.svg)](https://arxiv.org/abs/2604.21911)

3. **When Looking Is Not Enough: Visual Attention Structure Reveals Hallucination in MLLMs**  ,   Cao et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2605.11559-b31b1b.svg)](https://arxiv.org/abs/2605.11559)

4. **What Makes LVLMs Hallucinate Less? Unveiling the Architectural Factors Behind Hallucination Robustness**  ,   He et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2605.30911-b31b1b.svg)](https://arxiv.org/abs/2605.30911)

5. **Vocabulary Hijacking in LVLMs: Unveiling Critical Attention Heads by Excluding Inert Tokens to Mitigate Hallucination**  ,   Chen et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2605.10622-b31b1b.svg)](https://arxiv.org/abs/2605.10622)

6. **Visual Multi-Agent System: Mitigating Hallucination Snowballing via Visual Flow**  ,   Yu et al.  ,
7. **Vision-Language Introspection: Mitigating Overconfident Hallucinations in MLLMs via Interpretable Bi-Causal Steering**  ,   Liu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2601.05159-b31b1b.svg)](https://arxiv.org/abs/2601.05159)

8. **Vision-Language Introspection: Mitigating Overconfident Hallucinations in MLLMs via Interpretable Bi-Causal Steering**  ,   Liu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2601.05159-b31b1b.svg)](https://arxiv.org/abs/2601.05159)

9. **VIB-Probe: Detecting and Mitigating Hallucinations in Vision-Language Models via Variational Information Bottleneck**  ,   Zhang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2601.05547-b31b1b.svg)](https://arxiv.org/abs/2601.05547)

10. **Towards Interpretable Hallucination Analysis and Mitigation in LVLMs via Contrastive Neuron Steering**  ,   Lyu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.00621-b31b1b.svg)](https://arxiv.org/abs/2602.00621)

11. **Token-Guard: Towards Token-Level Hallucination Control via Self-Checking Decoding**  ,   Zhu et al.  ,
12. **Steering the Verifiability of Multimodal AI Hallucinations**  ,   Pang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.06714-b31b1b.svg)](https://arxiv.org/abs/2604.06714)

13. **Staying VIGILant: Mitigating Visual Laziness via Counterfactual Visual Alignment in MLLMs**  ,   Xiao et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2606.26387-b31b1b.svg)](https://arxiv.org/abs/2606.26387)

14. **Spotlight and Shadow: Attention-Guided Dual-Anchor Introspective Decoding for MLLM Hallucination Mitigation**  ,   Wu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.10071-b31b1b.svg)](https://arxiv.org/abs/2604.10071)

15. **SinkTrack: Attention Sink based Context Anchoring for Large Language Models**  ,   Liu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.10027-b31b1b.svg)](https://arxiv.org/abs/2604.10027)

16. **Seeing to Ground: Visual Attention for Hallucination-Resilient MDLLMs**  ,   Narnaware et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2603.25711-b31b1b.svg)](https://arxiv.org/abs/2603.25711)

17. **Seeing Right but Saying Wrong: Inter- and Intra-Layer Refinement in MLLMs without Training**  ,   Song et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2601.07359-b31b1b.svg)](https://arxiv.org/abs/2601.07359)

18. **See It, Say It, Sorted: An Iterative Training-Free Framework for Visually-Grounded Multimodal Reasoning in LVLMs**  ,   Zhang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.21497-b31b1b.svg)](https://arxiv.org/abs/2602.21497)

19. **See Fair, Speak Truth: Equitable Attention Improves Grounding and Reduces Hallucination in Vision-Language Alignment**  ,   Azeez et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.09749-b31b1b.svg)](https://arxiv.org/abs/2604.09749)

20. **SchröMind: Mitigating Hallucinations in Multimodal Large Language Models via Solving the Schrödinger Bridge Problem**  ,   Shi et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.09528-b31b1b.svg)](https://arxiv.org/abs/2602.09528)

21. **Scalpel: Fine-Grained Alignment of Attention Activation Manifolds via Mixture Gaussian Bridges to Mitigate Multimodal Hallucination**  ,   Shi et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.09541-b31b1b.svg)](https://arxiv.org/abs/2602.09541)

22. **STEAR: Layer-Aware Spatiotemporal Evidence Intervention for Hallucination Mitigation in Video Large Language Models**  ,   Fan et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.03045-b31b1b.svg)](https://arxiv.org/abs/2604.03045)

23. **SHIELD: Suppressing Hallucinations In LVLM Encoders via Bias and Vulnerability Defense**  ,   Huang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2510.16596-b31b1b.svg)](https://arxiv.org/abs/2510.16596)

24. **SDCD: Structure-Disrupted Contrastive Decoding for Mitigating Hallucinations in Large Vision-Language Models**  ,   Xia et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2601.03500-b31b1b.svg)](https://arxiv.org/abs/2601.03500)

25. **SAKED: Mitigating Hallucination in Large Vision-Language Models via Stability-Aware Knowledge Enhanced Decoding**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.09825-b31b1b.svg)](https://arxiv.org/abs/2602.09825)

26. **Revis: Sparse Latent Steering to Mitigate Object Hallucination in Large Vision-Language Models**  ,   Wu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.11824-b31b1b.svg)](https://arxiv.org/abs/2602.11824)

27. **Residual Decoding: Mitigating Hallucinations in Large Vision-Language Models via History-Aware Residual Guidance**  ,   Chen et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.01047-b31b1b.svg)](https://arxiv.org/abs/2602.01047)

28. **R-CoV: Region-Aware Chain-of-Verification for Alleviating Object Hallucinations in LVLMs**  ,   Xie et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.20696-b31b1b.svg)](https://arxiv.org/abs/2604.20696)

29. **Prefill-Time Intervention for Mitigating Hallucination in Large Vision-Language Models**  ,   Zhang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.25642-b31b1b.svg)](https://arxiv.org/abs/2604.25642)

30. **Overthinking Causes Hallucination: Tracing Confounder Propagation in Vision Language Models**  ,   Shoby et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2603.07619-b31b1b.svg)](https://arxiv.org/abs/2603.07619)

31. **One Token, Two Fates: A Unified Framework via Vision Token Manipulation Against MLLMs Hallucination**  ,   Fa et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2603.10360-b31b1b.svg)](https://arxiv.org/abs/2603.10360)

32. **On the Nature of Attention Sink that Shapes Decoding Strategy in Omni-LLMs**  ,   Yoo et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2603.14337-b31b1b.svg)](https://arxiv.org/abs/2603.14337)

33. **ORSc: Object-Aware Reinforcement with Semantic Consistency for Hallucination Mitigation in MLLMs**  ,   He et al.  ,   [Paper](https://ieeexplore.ieee.org/document/11464193/)
34. **NoLan: Mitigating Object Hallucinations in Large Vision-Language Models via Dynamic Suppression of Language Priors**  ,   Ren et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.22144-b31b1b.svg)](https://arxiv.org/abs/2602.22144)

35. **MoD-DPO: Towards Mitigating Cross-modal Hallucinations in Omni LLMs using Modality Decoupled Preference Optimization**  ,   Chaubey et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2603.03192-b31b1b.svg)](https://arxiv.org/abs/2603.03192)

36. **Mitigating Object and Relationship Hallucination in Large Vision Language Model with Multi-Agent Guidance**  ,   Kim et al.  ,   [Paper](https://ieeexplore.ieee.org/document/11463505/)
37. **Mitigating Object Hallucinations in Vision-Language Models through Region-Aware Attention Recalibration**  ,   Xu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2605.24957-b31b1b.svg)](https://arxiv.org/abs/2605.24957)

38. **Mitigating Object Hallucinations in LVLMs via Attention Imbalance Rectification**  ,   Sun et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2603.24058-b31b1b.svg)](https://arxiv.org/abs/2603.24058)

39. **Mitigating Hallucinations in Large Vision-Language Models via Causal Route Gating**  ,   Cheng et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2605.24024-b31b1b.svg)](https://arxiv.org/abs/2605.24024)

40. **Mitigating Entangled Steering in Large Vision-Language Models for Hallucination Reduction**  ,   Zhang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.07914-b31b1b.svg)](https://arxiv.org/abs/2604.07914)

41. **Mitigating Action-Relation Hallucinations in LVLMs via Relation-aware Visual Enhancement**  ,   Qin et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2605.11808-b31b1b.svg)](https://arxiv.org/abs/2605.11808)

42. **Mask What Matters: Mitigating Object Hallucinations in Multimodal Large Language Models with Object-Aligned Visual Contrastive Decoding**  ,   Chen et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.11737-b31b1b.svg)](https://arxiv.org/abs/2602.11737)

43. **MHSA: A Lightweight Framework for Mitigating Hallucinations via Steered Attention in LVLMs**  ,   Ding et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2605.14966-b31b1b.svg)](https://arxiv.org/abs/2605.14966)

44. **Lyapunov Probes for Hallucination Detection in Large Foundation Models**  ,   Luan et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2603.06081-b31b1b.svg)](https://arxiv.org/abs/2603.06081)

45. **Looking Back and Forth: Cross-Image Attention Calibration and Attentive Preference Learning for Multi-Image Hallucination Mitigation**  ,   Yang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2603.07048-b31b1b.svg)](https://arxiv.org/abs/2603.07048)

46. **Look Twice: Training-Free Evidence Highlighting in Multimodal Large Language Models**  ,   Morini et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.01280-b31b1b.svg)](https://arxiv.org/abs/2604.01280)

47. **Look Carefully: Adaptive Visual Reinforcements in Multimodal Large Language Models for Hallucination Mitigation**  ,   Zhu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.24041-b31b1b.svg)](https://arxiv.org/abs/2602.24041)

48. **Locate-then-Sparsify: Attribution Guided Sparse Strategy for Visual Hallucination Mitigation**  ,   Dang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2603.16284-b31b1b.svg)](https://arxiv.org/abs/2603.16284)

49. **KVSmooth: Mitigating Hallucination in Multi-modal Large Language Models through Key-Value Smoothing**  ,   Jiang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.04268-b31b1b.svg)](https://arxiv.org/abs/2602.04268)

50. **INFACT: A Diagnostic Benchmark for Induced Faithfulness and Factuality Hallucinations in Video-LLMs**  ,   Yang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2603.11481-b31b1b.svg)](https://arxiv.org/abs/2603.11481)

51. **HaloProbe: Bayesian Detection and Mitigation of Object Hallucinations in Vision-Language Models**  ,   Zohrabi et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.06165-b31b1b.svg)](https://arxiv.org/abs/2604.06165)

52. **Hallucination-aware intermediate representation edit in large vision-language models**  ,   Suo et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2603.29405-b31b1b.svg)](https://arxiv.org/abs/2603.29405)

53. **Hallucination Begins Where Saliency Drops**  ,   Zhang et al.  ,
54. **HalluClear: Diagnosing, Evaluating and Mitigating Hallucinations in GUI Agents**  ,   Jin et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.17284-b31b1b.svg)](https://arxiv.org/abs/2604.17284)

55. **HTDC: Hesitation-Triggered Differential Calibration for Mitigating Hallucination in Large Vision-Language Models**  ,   Liu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.12115-b31b1b.svg)](https://arxiv.org/abs/2604.12115)

56. **HIME: Mitigating Object Hallucinations in LVLMs via Hallucination Insensitivity Model Editing**  ,   Akl et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.18711-b31b1b.svg)](https://arxiv.org/abs/2602.18711)

57. **GroundCount: Grounding Vision-Language Models with Object Detection for Mitigating Counting Hallucinations**  ,   Chen et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2603.10978-b31b1b.svg)](https://arxiv.org/abs/2603.10978)

58. **GEASS: Training-Free Caption Steering for Hallucination Mitigation in Vision-Language Models**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2605.01733-b31b1b.svg)](https://arxiv.org/abs/2605.01733)

59. **Focus Matters: Phase-Aware Suppression for Hallucination in Vision-Language Models**  ,   Kim et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.03556-b31b1b.svg)](https://arxiv.org/abs/2604.03556)

60. **First Logit Boosting: Visual Grounding Method to Mitigate Object Hallucination in Large Vision-Language Models**  ,   Ha et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.00455-b31b1b.svg)](https://arxiv.org/abs/2604.00455)

61. **Finding the Correct Visual Evidence Without Forgetting: Mitigating Hallucination in LVLMs via Inter-Layer Visual Attention Discrepancy**  ,   Xie et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2605.20965-b31b1b.svg)](https://arxiv.org/abs/2605.20965)

62. **Dynamic Multimodal Activation Steering for Hallucination Mitigation in Large Vision-Language Models**  ,   Yin et al.  ,
63. **Dynamic Multimodal Activation Steering for Hallucination Mitigation in Large Vision-Language Models**  ,   Yin et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.21704-b31b1b.svg)](https://arxiv.org/abs/2602.21704)

64. **Do We Really Need External Tools to Mitigate Hallucinations? SIRA: Shared-Prefix Internal Reconstruction of Attribution**  ,   Qin et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2605.14621-b31b1b.svg)](https://arxiv.org/abs/2605.14621)

65. **Do I Really Know? Learning Factual Self-Verification for Hallucination Reduction**  ,   Altinisik et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.02018-b31b1b.svg)](https://arxiv.org/abs/2602.02018)

66. **Distorted or Fabricated? A Survey on Hallucination in Video LLMs**  ,   Huang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.12944-b31b1b.svg)](https://arxiv.org/abs/2604.12944)

67. **Detecting Contextual Hallucinations in LLMs with Frequency-Aware Attention**  ,   Qi et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.18145-b31b1b.svg)](https://arxiv.org/abs/2602.18145)

68. **Detecting Contextual Hallucinations in LLMs with Frequency-Aware Attention**  ,   Qi et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.18145-b31b1b.svg)](https://arxiv.org/abs/2602.18145)

69. **DetailVerifyBench: A Benchmark for Dense Hallucination Localization in Long Image Captions**  ,   Wang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.05623-b31b1b.svg)](https://arxiv.org/abs/2604.05623)

70. **Decoding by Perturbation: Mitigating MLLM Hallucinations via Dynamic Textual Perturbation**  ,   Jia et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.12424-b31b1b.svg)](https://arxiv.org/abs/2604.12424)

71. **Countering the Over-Reliance Trap: Mitigating Object Hallucination for LVLMs via a Self-Validation Framework**  ,   Liu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2601.22451-b31b1b.svg)](https://arxiv.org/abs/2601.22451)

72. **Correcting Visual Blur Induced by Attention Distraction to Reduce Hallucinations: Algorithm and Theory**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2605.24602-b31b1b.svg)](https://arxiv.org/abs/2605.24602)

73. **Copy-Paste to Mitigate Large Language Model Hallucinations**  ,   Long et al.  ,
74. **Causal Decoding for Hallucination-Resistant Multimodal Large Language Models**  ,   Tan et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.21441-b31b1b.svg)](https://arxiv.org/abs/2602.21441)

75. **CRoPS: A Training-Free Hallucination Mitigation Framework for Vision-Language Models**  ,   Anand et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2601.00659-b31b1b.svg)](https://arxiv.org/abs/2601.00659)

76. **CAST: Mitigating Object Hallucination in Large Vision-Language Models via Caption-Guided Visual Attention Steering**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2605.04641-b31b1b.svg)](https://arxiv.org/abs/2605.04641)

77. **Beyond the Global Scores: Fine-Grained Token Grounding as a Robust Detector of LVLM Hallucinations**  ,   Nguyen et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.04863-b31b1b.svg)](https://arxiv.org/abs/2604.04863)

78. **Beyond Static Cropping: Layer-Adaptive Visual Localization and Decoding Enhancement**  ,   Zhu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.04304-b31b1b.svg)](https://arxiv.org/abs/2602.04304)

79. **Beyond Dominant Patches: Spatial Credit Redistribution For Grounded Vision-Language Models**  ,   Samin et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.22469-b31b1b.svg)](https://arxiv.org/abs/2602.22469)

80. **Benchmarking Deflection and Hallucination in Large Vision-Language Models**  ,   Moratelli et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.12033-b31b1b.svg)](https://arxiv.org/abs/2604.12033)

81. **Attention-space Contrastive Guidance for Efficient Hallucination Mitigation in LVLMs**  ,   Jo et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2601.13707-b31b1b.svg)](https://arxiv.org/abs/2601.13707)

82. **Attention at Rest Stays at Rest: Breaking Visual Inertia for Cognitive Hallucination Mitigation**  ,   Gong et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.01989-b31b1b.svg)](https://arxiv.org/abs/2604.01989)

83. **Aligning What Vision-Language Models See and Perceive with Adaptive Information Flow**  ,   Liu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.15809-b31b1b.svg)](https://arxiv.org/abs/2604.15809)

84. **AdaVBoost: Mitigating Hallucinations in LVLMs via Token-Level Adaptive Visual Attention Boosting**  ,   Zhang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2602.13600-b31b1b.svg)](https://arxiv.org/abs/2602.13600)

85. **AdaIAT: Adaptively Increasing Attention to Generated Text to Alleviate Hallucinations in LVLM**  ,   Zhong et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2603.04908-b31b1b.svg)](https://arxiv.org/abs/2603.04908)

86. **AFTER: Mitigating the Object Hallucination of LVLM via Adaptive Factual-Guided Activation Editing**  ,   Wang et al.  ,
87. **ACT Now: Preempting LVLM Hallucinations via Adaptive Context Integration**  ,   Yan et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.00983-b31b1b.svg)](https://arxiv.org/abs/2604.00983)

88. **3D-VCD: Hallucination Mitigation in 3D-LLM Embodied Agents through Visual Contrastive Decoding**  ,   Ogunleye et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2604.08645-b31b1b.svg)](https://arxiv.org/abs/2604.08645)

## 2025
89. **Why LVLMs Are More Prone to Hallucinations in Longer Responses: The Role of Context**  ,   Zheng et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2510.20229-b31b1b.svg)](https://arxiv.org/abs/2510.20229)

90. **When Semantics Mislead Vision: Mitigating Large Multimodal Models Hallucinations in Scene Text Spotting and Understanding**  ,   Shu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2506.05551-b31b1b.svg)](https://arxiv.org/abs/2506.05551)

91. **When Images Speak Louder: Mitigating Language Bias-induced Hallucinations in VLMs through Cross-Modal Guidance**  ,   Cao et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2510.10466-b31b1b.svg)](https://arxiv.org/abs/2510.10466)

92. **Watch Closely: Mitigating Object Hallucinations in Large Vision-Language Models with Disentangled Decoding**  ,   Ma et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2512.19070-b31b1b.svg)](https://arxiv.org/abs/2512.19070)

93. **Visual Evidence Prompting Mitigates Hallucinations in Large Vision-Language Models**  ,   Li et al.  ,
94. **Visual Description Grounding Reduces Hallucinations and Boosts Reasoning in LVLMs**  ,   Ghosh et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2405.15683-b31b1b.svg)](https://arxiv.org/abs/2405.15683)

95. **Visual Attention Never Fades: Selective Progressive Attention ReCalibration for Detailed Image Captioning in Multimodal Large Language Models**  ,   Jung et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2502.01419-b31b1b.svg)](https://arxiv.org/abs/2502.01419)

96. **VaLiD: Mitigating the Hallucination of Large Vision Language Models by Visual Layer Fusion Contrastive Decoding**  ,   Wang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2411.15839-b31b1b.svg)](https://arxiv.org/abs/2411.15839)

97. **VEGAS: Mitigating Hallucinations in Large Vision-Language Models via Vision-Encoder Attention Guided Adaptive Steering**  ,   Wang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2512.12089-b31b1b.svg)](https://arxiv.org/abs/2512.12089)

98. **VASparse: Towards Efficient Visual Hallucination Mitigation via Visual-Aware Token Sparsification**  ,   Zhuang et al.  ,
99. **VASparse: Towards Efficient Visual Hallucination Mitigation via Visual-Aware Token Sparsification**  ,   Zhuang et al.  ,
100. **Unveiling the Response of Large Vision-Language Models to Visually Absent Tokens**  ,   Kim et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2509.03025-b31b1b.svg)](https://arxiv.org/abs/2509.03025)

101. **Understanding and Mitigating Hallucination in Large Vision-Language Models via Modular Attribution and Intervention**  ,   Yang et al.  ,
102. **TruthPrInt: Mitigating LVLM Object Hallucination Via Latent Truthful-Guided Pre-Intervention**  ,   Duan et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2503.10602-b31b1b.svg)](https://arxiv.org/abs/2503.10602)

103. **Toward Robust Hyper-Detailed Image Captioning: A Multiagent Approach and Dual Evaluation Metrics for Factuality and Coverage**  ,   Lee et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2412.15484-b31b1b.svg)](https://arxiv.org/abs/2412.15484)

104. **To Sink or Not to Sink: Visual Information Pathways in Large Vision-Language Models**  ,   Luo et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2510.08510-b31b1b.svg)](https://arxiv.org/abs/2510.08510)

105. **Through the Magnifying Glass: Adaptive Perception Magnification for Hallucination-Free VLM Decoding**  ,   Mao et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2503.10183-b31b1b.svg)](https://arxiv.org/abs/2503.10183)

106. **Tell Model Where to Look: Mitigating Hallucinations in MLLMs by Vision-Guided Attention**  ,   Zhao et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2511.20032-b31b1b.svg)](https://arxiv.org/abs/2511.20032)

107. **TPC: Cross-Temporal Prediction Connection for Vision-Language Model Hallucination Reduction**  ,   Wang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2503.04457-b31b1b.svg)](https://arxiv.org/abs/2503.04457)

108. **TARS: MinMax Token-Adaptive Preference Strategy for MLLM Hallucination Reduction**  ,   Zhang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2507.21584-b31b1b.svg)](https://arxiv.org/abs/2507.21584)

109. **TARAC: Mitigating Hallucination in LVLMs via Temporal Attention Real-time Accumulative Connection**  ,   Xie et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2504.04099-b31b1b.svg)](https://arxiv.org/abs/2504.04099)

110. **Suppressing VLM Hallucinations with Spectral Representation Filtering**  ,   Ali et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2511.12220-b31b1b.svg)](https://arxiv.org/abs/2511.12220)

111. **Steering LVLMs via Sparse Autoencoder for Hallucination Mitigation**  ,   Hua et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2505.16146-b31b1b.svg)](https://arxiv.org/abs/2505.16146)

112. **Shallow Focus, Deep Fixes: Enhancing Shallow Layers Vision Attention Sinks to Alleviate Hallucination in LVLMs**  ,   Zhang et al.  ,
113. **Self-Introspective Decoding: Alleviating Hallucinations for Large Vision-Language Models**  ,   Huo et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2408.02032-b31b1b.svg)](https://arxiv.org/abs/2408.02032)

114. **Self-Correcting Decoding with Generative Feedback for Mitigating Hallucinations in Large Vision-Language Models**  ,   Zhang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2502.06130-b31b1b.svg)](https://arxiv.org/abs/2502.06130)

115. **Self-Consistency as a Free Lunch: Reducing Hallucinations in Vision-Language Models via Self-Reflection**  ,   Han et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2509.23236-b31b1b.svg)](https://arxiv.org/abs/2509.23236)

116. **Self-Augmented Visual Contrastive Decoding**  ,   Im et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2510.13315-b31b1b.svg)](https://arxiv.org/abs/2510.13315)

117. **Seeing but Not Believing: Probing the Disconnect Between Visual Attention and Answer Correctness in VLMs**  ,   Liu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2510.17771-b31b1b.svg)](https://arxiv.org/abs/2510.17771)

118. **Seeing It or Not? Interpretable Vision-aware Latent Steering to Mitigate Object Hallucinations**  ,   Chen et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2505.17812-b31b1b.svg)](https://arxiv.org/abs/2505.17812)

119. **Seeing Far and Clearly: Mitigating Hallucinations in MLLMs with Attention Causal Decoding**  ,   Tang et al.  ,
120. **See What You Are Told: Visual Attention Sink in Large Multimodal Models**  ,   Kang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2503.03321-b31b1b.svg)](https://arxiv.org/abs/2503.03321)

121. **SECOND: Mitigating Perceptual Hallucination in Vision-Language Models via Selective and Contrastive Decoding**  ,   Park et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2506.08391-b31b1b.svg)](https://arxiv.org/abs/2506.08391)

122. **SECOND: Mitigating Perceptual Hallucination in Vision-Language Models via Selective and Contrastive Decoding**  ,   Park et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2506.08391-b31b1b.svg)](https://arxiv.org/abs/2506.08391)

123. **SAVER: Mitigating Hallucinations in Large Vision-Language Models via Style-Aware Visual Early Revision**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2508.03177-b31b1b.svg)](https://arxiv.org/abs/2508.03177)

124. **SAVE: Sparse Autoencoder-Driven Visual Information Enhancement for Mitigating Object Hallucination**  ,   Park et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2512.07730-b31b1b.svg)](https://arxiv.org/abs/2512.07730)

125. **Revisit What You See: Disclose Language Prior in Vision Tokens for Efficient Guided Decoding of LVLMs**  ,   Cho et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2506.09522-b31b1b.svg)](https://arxiv.org/abs/2506.09522)

126. **Revealing Perception and Generation Dynamics in LVLMs: Mitigating Hallucinations via Validated Dominance Correction**  ,   Lyu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2512.18813-b31b1b.svg)](https://arxiv.org/abs/2512.18813)

127. **Reducing Hallucinations in Vision-Language Models via Latent Space Steering**  ,   Liu et al.  ,
128. **Re-Align: Aligning Vision Language Models via Retrieval-Augmented Direct Preference Optimization**  ,   Xing et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2502.13146-b31b1b.svg)](https://arxiv.org/abs/2502.13146)

129. **RAR: Reversing Visual Attention Re-sinking for Unlocking Potential in Multimodal Large Language Models**  ,   Anonymous et al.  ,
130. **Qwen Look Again: Guiding Vision-Language Reasoning Models to Re-attention Visual Information**  ,   Chu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2505.23558-b31b1b.svg)](https://arxiv.org/abs/2505.23558)

131. **Poison as Cure: Visual Noise for Mitigating Object Hallucinations in LVMs**  ,   Zhang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2501.19164-b31b1b.svg)](https://arxiv.org/abs/2501.19164)

132. **PerturboLLaVA: Reducing Multimodal Hallucinations with Perturbative Visual Training**  ,   Chen et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2503.06486-b31b1b.svg)](https://arxiv.org/abs/2503.06486)

133. **PerturboLLaVA: Reducing Multimodal Hallucinations with Perturbative Visual Training**  ,   Chen et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2503.06486-b31b1b.svg)](https://arxiv.org/abs/2503.06486)

134. **Paying More Attention to Image: A Training-Free Method for Alleviating Hallucination in LVLMs**  ,   Anonymous et al.  ,   [Paper](https://link.springer.com/10.1007/978-3-031-73010-8_8)
135. **PAINT: Paying Attention to INformed Tokens to Mitigate Hallucination in Large Vision-Language Model**  ,   Arif et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2501.12206-b31b1b.svg)](https://arxiv.org/abs/2501.12206)

136. **Octopus: Alleviating Hallucination via Dynamic Contrastive Decoding**  ,   Suo et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2503.00361-b31b1b.svg)](https://arxiv.org/abs/2503.00361)

137. **ONLY: One-Layer Intervention Sufficiently Mitigates Hallucinations in Large Vision-Language Models**  ,   Wan et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2507.00898-b31b1b.svg)](https://arxiv.org/abs/2507.00898)

138. **ONLY: One-Layer Intervention Sufficiently Mitigates Hallucinations in Large Vision-Language Models**  ,   Wan et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2507.00898-b31b1b.svg)](https://arxiv.org/abs/2507.00898)

139. **Nullu: Mitigating Object Hallucinations in Large Vision-Language Models via HalluSpace Projection**  ,   Yang et al.  ,
140. **Not All Tokens and Heads Are Equally Important: Dual-Level Attention Intervention for Hallucination Mitigation**  ,   Tang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2506.12609-b31b1b.svg)](https://arxiv.org/abs/2506.12609)

141. **MoLE: Decoding by Mixture of Layer Experts Alleviates Hallucination in Large Vision-Language Models**  ,   Liang et al.  ,
142. **Mixture of Decoding: An Attention-Inspired Adaptive Decoding Strategy to Mitigate Hallucinations in Large Vision-Language Models**  ,   Chen et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2505.17061-b31b1b.svg)](https://arxiv.org/abs/2505.17061)

143. **Mitigating Object Hallucinations via Sentence-Level Early Intervention**  ,   Peng et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2507.12455-b31b1b.svg)](https://arxiv.org/abs/2507.12455)

144. **Mitigating Object Hallucinations in MLLMs via Multi-Frequency Perturbations**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2503.14895-b31b1b.svg)](https://arxiv.org/abs/2503.14895)

145. **Mitigating Object Hallucinations in Large Vision-Language Models with Assembly of Global and Local Attention**  ,   An et al.  ,
146. **Mitigating Object Hallucinations in Large Vision-Language Models via Attention Calibration**  ,   Zhu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2502.01969-b31b1b.svg)](https://arxiv.org/abs/2502.01969)

147. **Mitigating Object Hallucinations in Large Vision-Language Models via Attention Calibration**  ,   Zhu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2502.01969-b31b1b.svg)](https://arxiv.org/abs/2502.01969)

148. **Mitigating Object Hallucination in Large Vision-Language Models via Image-Grounded Guidance**  ,   Zhao et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2402.08680-b31b1b.svg)](https://arxiv.org/abs/2402.08680)

149. **Mitigating Modality Prior-Induced Hallucinations in Multimodal Large Language Models via Deciphering Attention Causality**  ,   Zhou et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2410.04780-b31b1b.svg)](https://arxiv.org/abs/2410.04780)

150. **Mitigating Hallucinations in Multi-modal Large Language Models via Image Token Attention-Guided Decoding**  ,   Xu et al.  ,
151. **Mitigating Hallucinations in Large Vision-Language Models via Reasoning Uncertainty-Guided Refinement**  ,   Li et al.  ,   [Paper](https://ieeexplore.ieee.org/document/11125489/)
152. **Mitigating Hallucinations in Large Vision-Language Models via Entity-Centric Multimodal Preference Optimization**  ,   Wu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2506.04039-b31b1b.svg)](https://arxiv.org/abs/2506.04039)

153. **Mitigating Hallucinations in Large Vision-Language Models by Adaptively Constraining Information Flow**  ,   Bai et al.  ,
154. **Mitigating Hallucination in Multimodal LLMs with Layer Contrastive Decoding**  ,   Tong et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2509.25177-b31b1b.svg)](https://arxiv.org/abs/2509.25177)

155. **Mitigating Hallucination in Large Vision-Language Models via Adaptive Attention Calibration**  ,   Fazli et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2505.21472-b31b1b.svg)](https://arxiv.org/abs/2505.21472)

156. **Mitigating Hallucination for Large Vision Language Model by Inter-Modality Correlation Calibration Decoding**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2501.01926-b31b1b.svg)](https://arxiv.org/abs/2501.01926)

157. **Mitigating Hallucination for Large Vision Language Model by Inter-Modality Correlation Calibration Decoding**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2501.01926-b31b1b.svg)](https://arxiv.org/abs/2501.01926)

158. **Mitigating Attention Sinks and Massive Activations in Audio-Visual Speech Recognition with LLMs**  ,   Anand et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2510.22603-b31b1b.svg)](https://arxiv.org/abs/2510.22603)

159. **Mitigate Language Priors in Large Vision-Language Models by Cross-Images Contrastive Decoding**  ,   Zhao et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2505.10634-b31b1b.svg)](https://arxiv.org/abs/2505.10634)

160. **MaskCD: Mitigating LVLM Hallucinations by Image Head Masked Contrastive Decoding**  ,   Deng et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2510.02790-b31b1b.svg)](https://arxiv.org/abs/2510.02790)

161. **MLLMs Know Where to Look: Training-free Perception of Small Visual Details with Multimodal LLMs**  ,   Zhang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2502.17422-b31b1b.svg)](https://arxiv.org/abs/2502.17422)

162. **MLLM can see? Dynamic Correction Decoding for Hallucination Mitigation**  ,   Wang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2410.11779-b31b1b.svg)](https://arxiv.org/abs/2410.11779)

163. **MINT: Mitigating Hallucinations in Large Vision-Language Models via Token Reduction**  ,   Wang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2502.00717-b31b1b.svg)](https://arxiv.org/abs/2502.00717)

164. **MIHBench: Benchmarking and Mitigating Multi-Image Hallucinations in Multimodal Large Language Models**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2508.00726-b31b1b.svg)](https://arxiv.org/abs/2508.00726)

165. **MCA-LLaVA: Manhattan Causal Attention for Reducing Hallucination in Large Vision-Language Models**  ,   Zhao et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2507.09184-b31b1b.svg)](https://arxiv.org/abs/2507.09184)

166. **MAP: Mitigating Hallucinations in Large Vision-Language Models with Map-Level Attention Processing**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2508.01653-b31b1b.svg)](https://arxiv.org/abs/2508.01653)

167. **Look Twice Before You Answer: Memory-Space Visual Retracing for Hallucination Mitigation in Multimodal Large Language Models**  ,   Zou et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2410.03577-b31b1b.svg)](https://arxiv.org/abs/2410.03577)

168. **Intervening Anchor Token: Decoding Strategy in Alleviating Hallucinations for MLLMs**  ,   Tang et al.  ,
169. **Interpreting and Editing Vision-Language Representations to Mitigate Hallucinations**  ,   Jiang et al.  ,
170. **Interpreting and Editing Vision-Language Representations to Mitigate Hallucinations**  ,   Jiang et al.  ,
171. **Insight Over Sight: Exploring the Vision-Knowledge Conflicts in Multimodal LLMs**  ,   Liu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2410.08145-b31b1b.svg)](https://arxiv.org/abs/2410.08145)

172. **InEx: Hallucination Mitigation via Introspection and Cross-Modal Multi-Agent Collaboration**  ,   Yang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2512.02981-b31b1b.svg)](https://arxiv.org/abs/2512.02981)

173. **Improve Decoding Factuality by Token-wise Cross Layer Entropy of Large Language Models**  ,   Wu et al.  ,
174. **INTER: Mitigating Hallucination in Large Vision-Language Models by Interaction Guidance Sampling**  ,   Dong et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2507.05056-b31b1b.svg)](https://arxiv.org/abs/2507.05056)

175. **Hallucinatory Image Tokens: A Training-free EAZY Approach on Detecting and Mitigating Object Hallucinations in LVLMs**  ,   Che et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2503.07772-b31b1b.svg)](https://arxiv.org/abs/2503.07772)

176. **Grounding Language with Vision: A Conditional Mutual Information Calibrated Decoding Strategy for Reducing Hallucinations in LVLMs**  ,   Fang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2505.19678-b31b1b.svg)](https://arxiv.org/abs/2505.19678)

177. **GHOST: Hallucination-Inducing Image Generation for Multimodal LLMs**  ,   Parast et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2509.25178-b31b1b.svg)](https://arxiv.org/abs/2509.25178)

178. **Focus on What Matters: Enhancing Medical Vision-Language Models with Automatic Attention Alignment Tuning**  ,   Chang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2505.18503-b31b1b.svg)](https://arxiv.org/abs/2505.18503)

179. **Extracting Visual Facts from Intermediate Layers for Mitigating Hallucinations in Multimodal Large Language Models**  ,   Zhou et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2507.15652-b31b1b.svg)](https://arxiv.org/abs/2507.15652)

180. **Enhancing Visual Reliance in Text Generation: A Bayesian Perspective on Mitigating Hallucination in Large Vision-Language Models**  ,   Hu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2505.19498-b31b1b.svg)](https://arxiv.org/abs/2505.19498)

181. **Energy-Guided Decoding for Object Hallucination Mitigation**  ,   Liu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2507.07731-b31b1b.svg)](https://arxiv.org/abs/2507.07731)

182. **Don't Miss the Forest for the Trees: Attentional Vision Calibration for Large Vision Language Models**  ,   Woo et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2405.17820-b31b1b.svg)](https://arxiv.org/abs/2405.17820)

183. **Don't Deceive Me: Mitigating Gaslighting through Attention Reallocation in LMMs**  ,   Jiao et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2504.09456-b31b1b.svg)](https://arxiv.org/abs/2504.09456)

184. **Devils in Middle Layers of Large Vision-Language Models: Interpreting, Detecting and Mitigating Object Hallucinations via Attention Lens**  ,   Jiang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2411.16724-b31b1b.svg)](https://arxiv.org/abs/2411.16724)

185. **Decoupling Contrastive Decoding: Robust Hallucination Mitigation in Multimodal Large Language Models**  ,   Chen et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2504.08809-b31b1b.svg)](https://arxiv.org/abs/2504.08809)

186. **Damo: Decoding by Accumulating Activations Momentum for Mitigating Hallucinations in Vision-Language Models**  ,   Wang et al.  ,
187. **D-LEAF: Localizing and Correcting Hallucinations in Multimodal LLMs via Layer-to-head Attention Diagnostics**  ,   Yang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2509.07864-b31b1b.svg)](https://arxiv.org/abs/2509.07864)

188. **Cross-Image Contrastive Decoding: Precise, Lossless Suppression of Language Priors in Large Vision-Language Models**  ,   Zhao et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2505.10634-b31b1b.svg)](https://arxiv.org/abs/2505.10634)

189. **Cracking the Code of Hallucination in LVLMs with Vision-aware Head Divergence**  ,   He et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2412.13949-b31b1b.svg)](https://arxiv.org/abs/2412.13949)

190. **Cracking the Code of Hallucination in LVLMs with Vision-aware Head Divergence**  ,   He et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2412.13949-b31b1b.svg)](https://arxiv.org/abs/2412.13949)

191. **Conscious Gaze: Adaptive Attention Mechanisms for Hallucination Mitigation in Vision-Language Models**  ,   Bu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2512.05546-b31b1b.svg)](https://arxiv.org/abs/2512.05546)

192. **CoFi-Dec: Hallucination-Resistant Decoding via Coarse-to-Fine Generative Feedback in Large Vision-Language Models**  ,   Cao et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2512.23453-b31b1b.svg)](https://arxiv.org/abs/2512.23453)

193. **ClearSight: Visual Signal Enhancement for Object Hallucination Mitigation in Multimodal Large Language Models**  ,   Yin et al.  ,
194. **Causally-Grounded Dual-Path Attention Intervention for Object Hallucination Mitigation in LVLMs**  ,   Yu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2511.09018-b31b1b.svg)](https://arxiv.org/abs/2511.09018)

195. **Causally-Grounded Dual-Path Attention Intervention for Object Hallucination Mitigation in LVLMs**  ,   Yu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2511.09018-b31b1b.svg)](https://arxiv.org/abs/2511.09018)

196. **Causal Tracing of Object Representations in Large Vision Language Models: Mechanistic Interpretability and Hallucination Mitigation**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2511.05923-b31b1b.svg)](https://arxiv.org/abs/2511.05923)

197. **Capturing Gaze Shifts for Guidance: Cross-Modal Fusion Enhancement for VLM Hallucination Mitigation**  ,   Qi et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2510.22067-b31b1b.svg)](https://arxiv.org/abs/2510.22067)

198. **CLAIM: Mitigating Multilingual Object Hallucination in Large Vision-Language Models with Cross-Lingual Attention Intervention**  ,   Ye et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2506.11073-b31b1b.svg)](https://arxiv.org/abs/2506.11073)

199. **CLAIM: Mitigating Multilingual Object Hallucination in Large Vision-Language Models with Cross-Lingual Attention Intervention**  ,   Ye et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2506.11073-b31b1b.svg)](https://arxiv.org/abs/2506.11073)

200. **CAI: Caption-Sensitive Attention Intervention for Mitigating Object Hallucination in Large Vision-Language Models**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2506.23590-b31b1b.svg)](https://arxiv.org/abs/2506.23590)

201. **CAI: Caption-Sensitive Attention Intervention for Mitigating Object Hallucination in Large Vision-Language Models**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2506.23590-b31b1b.svg)](https://arxiv.org/abs/2506.23590)

202. **CAAC: Confidence-Aware Attention Calibration to Reduce Hallucinations in Large Vision-Language Models**  ,   Fazli et al.  ,
203. **Beyond Single Models: Mitigating Multimodal Hallucinations via Adaptive Token Ensemble Decoding**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2510.18321-b31b1b.svg)](https://arxiv.org/abs/2510.18321)

204. **Benchmarking and Bridging Emotion Conflicts for Multimodal Emotion Reasoning**  ,   Han et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2508.01181-b31b1b.svg)](https://arxiv.org/abs/2508.01181)

205. **BIMA: Bijective Maximum Likelihood Learning Approach to Hallucination Prediction and Mitigation in Large Vision-Language Models**  ,   Tran et al.  ,
206. **Attention Hijackers: Detect and Disentangle Attention Hijacking in LVLMs for Hallucination Mitigation**  ,   Chen et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2503.08216-b31b1b.svg)](https://arxiv.org/abs/2503.08216)

207. **Analyzing and Mitigating Object Hallucination: A Training Bias Perspective**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2508.04567-b31b1b.svg)](https://arxiv.org/abs/2508.04567)

208. **Alleviating Hallucination in Large Vision-Language Models with Active Retrieval Augmentation**  ,   Qu et al.  ,
209. **Adaptive Residual-Update Steering for Low-Overhead Hallucination Mitigation in Large Vision Language Models**  ,   Zou et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2511.10292-b31b1b.svg)](https://arxiv.org/abs/2511.10292)

210. **Activation Steering Decoding: Mitigating Hallucination in Large Vision-Language Models through Bidirectional Hidden State Intervention**  ,   Su et al.  ,
211. **AVCD: Mitigating Hallucinations in Audio-Visual Large Language Models through Contrastive Decoding**  ,   Jung et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2505.20862-b31b1b.svg)](https://arxiv.org/abs/2505.20862)

212. **ASCD: Attention-Steerable Contrastive Decoding for Reducing Hallucination in MLLM**  ,   Wang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2506.14766-b31b1b.svg)](https://arxiv.org/abs/2506.14766)

213. **ASCD: Attention-Steerable Contrastive Decoding for Reducing Hallucination in MLLM**  ,   Wang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2506.14766-b31b1b.svg)](https://arxiv.org/abs/2506.14766)

214. **A Survey of Multimodal Hallucination Evaluation and Detection**  ,   Chen et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2507.19024-b31b1b.svg)](https://arxiv.org/abs/2507.19024)

## 2024
215. **verthinking Causes Hallucination: Tracing Confounder Propagation in Vision Language Models**  ,   Shoby et al.  ,
216. **ZINA: Multimodal Fine-grained Hallucination Detection and Editing**  ,   Wada et al.  ,
217. **Vocabulary Hijacking in LVLMs: Unveiling Critical Attention Heads by Excluding Inert Tokens to Mitigate Hallucination**  ,   Chen et al.  ,
218. **Vision-Language Introspection: Mitigating Overconfident Hallucinations in MLLMs via Interpretable Bi-Causal Steering**  ,   Liu et al.  ,
219. **Verb Mirage: Unveiling and Assessing Verb Concept Hallucinations in Multimodal Large Language Models**  ,   Wang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2412.04939-b31b1b.svg)](https://arxiv.org/abs/2412.04939)

220. **VaLiD: Mitigating the Hallucination of Large Vision Language Models by Visual Layer Fusion Contrastive Decoding**  ,   Wang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2411.15839-b31b1b.svg)](https://arxiv.org/abs/2411.15839)

221. **VIB-Probe: Detecting and Mitigating Hallucinations in Vision-Language Models via Variational Information Bottleneck**  ,   Zhang et al.  ,
222. **VES-RFT: Rewarding Visual Evidence Sensitivity to Mitigate Hallucinations in Large Vision-Language Models**  ,   Hou et al.  ,
223. **VACoDe: Visual Augmented Contrastive Decoding**  ,   Kim et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2408.05337-b31b1b.svg)](https://arxiv.org/abs/2408.05337)

224. **Unstitching the Chimera: Frame-Level Risk and Train-Free Mitigation for Video Hallucination**  ,   Yang et al.  ,
225. **Unraveling Cross-Modality Knowledge Conflict in Large Vision-Language Models**  ,   Zhu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2410.03659-b31b1b.svg)](https://arxiv.org/abs/2410.03659)

226. **Understanding the Role of Hallucination in Reinforcement Post-Training of Multimodal Reasoning Models**  ,   Zhang et al.  ,
227. **Understanding and Mitigating Hallucinations in Multimodal Chain-of-Thought Models**  ,   Ma et al.  ,
228. **Understanding and Mitigating Hallucinations in Multimodal Chain-of-Thought Models**  ,   Ma et al.  ,
229. **TriDF: Evaluating Perception, Detection, and Hallucination for Interpretable DeepFake Detection**  ,   Jiang-Lin et al.  ,
230. **Thinking in Uncertainty: Mitigating Hallucinations in MLRMs with Latent Entropy-Aware Decoding**  ,   Xu et al.  ,
231. **TELL YOUR MODEL WHERE TO ATTEND: POST-HOC ATTENTION STEERING FOR LLMS**  ,   Zhang et al.  ,
232. **Spotlight and Shadow: Attention-Guided Dual-Anchor Introspective Decoding for MLLM Hallucination Mitigation**  ,   Wu et al.  ,
233. **Seeing the Image: Prioritizing Visual Correlation by Contrastive Alignment**  ,   Xiao et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2405.17871-b31b1b.svg)](https://arxiv.org/abs/2405.17871)

234. **Seeing Clearly by Layer Two: Enhancing Attention Heads to Alleviate Hallucination in LVLMs**  ,   Zhang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2411.09968-b31b1b.svg)](https://arxiv.org/abs/2411.09968)

235. **Same Attention, Different Truths: Put Logit-Lens over Visual Attention to Detect and Mitigate LVLM Object Hallucination**  ,   Wang et al.  ,
236. **SVHalluc: Benchmarking Speech-Vision Hallucination in Audio-Visual Large Language Models**  ,   Zhang et al.  ,
237. **SEASON: Mitigating Temporal Hallucination in Video Large Language Models via Self-Diagnostic Contrastive Decoding**  ,   Wu et al.  ,
238. **Revealing and Enhancing Core Visual Regions: Harnessing Internal Attention Dynamics for Hallucination Mitigation in LVLMs**  ,   Lyu et al.  ,
239. **Residual Decoding: Mitigating Hallucinations in Large Vision-Language Models via History-Aware Residual Guidance**  ,   Chen et al.  ,
240. **RITUAL: Random Image Transformations as a Universal Anti-hallucination Lever in Large Vision Language Models**  ,   Woo et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2405.17821-b31b1b.svg)](https://arxiv.org/abs/2405.17821)

241. **Perceptual Hallucination in Vision–Language Models: Definition, Analysis and Verification**  ,   Hwang et al.  ,
242. **Once Correct, Still Wrong: Counterfactual Hallucination in Multilingual Vision-Language Models**  ,   Mousi et al.  ,
243. **OPERA: Alleviating Hallucination in Multi-Modal Large Language Models via Over-Trust Penalty and Retrospection-Allocation**  ,   Huang et al.  ,   [Paper](https://ieeexplore.ieee.org/document/10655465/)
244. **Multi-Modal Hallucination Control by Visual Information Grounding**  ,   Favero et al.  ,   [Paper](https://ieeexplore.ieee.org/document/10655750/)
245. **Mitigating Object Hallucination via Concentric Causal Attention**  ,   Xing et al.  ,
246. **Mitigating Object Hallucination in LVLMs via Attention Imbalance Rectification**  ,   Sun et al.  ,
247. **Mitigating Multimodal Hallucinations via Gradient-based Self-Reflection**  ,   Wang et al.  ,
248. **Mitigating Multilingual Hallucination in Large Vision-Language Models**  ,   Qu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2408.00550-b31b1b.svg)](https://arxiv.org/abs/2408.00550)

249. **Mitigating Hallucinations in VLMs: Enhancing Visual Attention via Head-Wise Perturbation**  ,   Wang et al.  ,
250. **Mitigating Hallucinations in Large Vision-Language Models without Performance Degradation**  ,   Zhu et al.  ,
251. **Mitigating Hallucinations in Large Vision-Language Models with Instruction Contrastive Decoding**  ,   Wang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2403.18715-b31b1b.svg)](https://arxiv.org/abs/2403.18715)

252. **Mitigating Action-Relation Hallucinations in LVLMs via Relation-aware Visual Enhancement**  ,   Qin et al.  ,
253. **MAD: Modality-Adaptive Decoding for Mitigating Cross-Modal Hallucinations in Multimodal Large Language Models**  ,   Chung et al.  ,
254. **Lyapunov Probes for Hallucination Detection in Large Foundation Models**  ,   Luan et al.  ,
255. **Layer Importance and Hallucination Analysis in Large Language Models via Enhanced Activation Variance-Sparsity**  ,   Song et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2411.10069-b31b1b.svg)](https://arxiv.org/abs/2411.10069)

256. **Latent Attention Denoising: A Training-Free Energy-Based Framework for Mitigating Hallucinations in Vision-Language Models**  ,   Luo et al.  ,
257. **Instructive Decoding: Instruction-Tuned Large Language Models are Self-Refiner from Noisy Instructions**  ,   Kim et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2311.00233-b31b1b.svg)](https://arxiv.org/abs/2311.00233)

258. **Inject to Heal: Alleviating hallucination in LVLMs via Context Embedding Injection**  ,   Fazli et al.  ,
259. **In-Context Sharpness as Alerts: An Inner Representation Perspective for Hallucination Mitigation**  ,   Chen et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2403.01548-b31b1b.svg)](https://arxiv.org/abs/2403.01548)

260. **IBD: Alleviating Hallucinations in Large Vision-Language Models via Image-Biased Decoding**  ,   Zhu et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2402.18476-b31b1b.svg)](https://arxiv.org/abs/2402.18476)

261. **HulluEdit: Single-Pass Evidence-Consistent Subspace Editing for Mitigating Hallucinations in Large Vision-Language Models**  ,   Lin et al.  ,
262. **HALC: Object Hallucination Reduction via Adaptive Focal-Contrast Decoding**  ,   Chen et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2403.00425-b31b1b.svg)](https://arxiv.org/abs/2403.00425)

263. **HAIT: Hybrid Adversarial Iterative Training for Mitigating Object Hallucination in Large Vision-Language Models**  ,   Zhao et al.  ,
264. **HAD: Hallucination-Aware Diffusion Priors for 3D Reconstruction**  ,   Liu et al.  ,
265. **Global Context or Local Detail? Adaptive Visual Grounding for Hallucination Mitigation**  ,   Jiang et al.  ,
266. **First Logit Boosting: Visual Grounding Method to Mitigate Object Hallucination in Large Vision-Language Models**  ,   Ha et al.  ,
267. **Fine-Grained Multi Image Object Hallucination Benchmark**  ,   Min et al.  ,
268. **Fighting Hallucinations with Counterfactuals: Diffusion-Guided Perturbations for LVLM Hallucination Suppression**  ,   Dastmalchi et al.  ,
269. **Envision, Attend, Then Respond: Counterfactual Hallucination Mitigation in Large Vision-Language Models**  ,   Liang et al.  ,
270. **ELV-Halluc: Benchmarking Semantic Aggregation Hallucinations in Video Understanding**  ,   Lu et al.  ,
271. **DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Models**  ,   Chuang et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2309.03883-b31b1b.svg)](https://arxiv.org/abs/2309.03883)

272. **DiVE: Decoupling Intra-layer Visual Evidence for Mitigating Hallucinations in Large Vision-Language Models**  ,   Li et al.  ,
273. **DAMRO: Dive into the Attention Mechanism of LVLM to Reduce Object Hallucination**  ,   Gong et al.  ,   [Paper](https://aclanthology.org/2024.emnlp-main.439)
274. **Cross-Modal Attention Calibration for LVLM Hallucination Mitigation**  ,   Li et al.  ,
275. **Counterfactual Segmentation Reasoning: Diagnosing and Mitigating Pixel-Grounding Hallucination**  ,   Li et al.  ,
276. **Contrastive Region Guidance: Improving Grounding in Vision-Language Models without Training**  ,   Wan et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2403.02325-b31b1b.svg)](https://arxiv.org/abs/2403.02325)

277. **Contrastive Decoding Reduces Hallucinations in Large Multilingual Machine Translation Models**  ,   Waldendorf et al.  ,
278. **ConVis: Contrastive Decoding with Hallucination Visualization for Mitigating Hallucinations in Multimodal Large Language Models**  ,   Park et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2408.13906-b31b1b.svg)](https://arxiv.org/abs/2408.13906)

279. **CausalLens: Sensitivity-Guided Multi-Head Causal Intervention for Hallucination Mitigation in Large Vision-Language Models**  ,   Ji et al.  ,
280. **COPO: Causal-Oriented Policy Optimization for Hallucinations of MLLMs**  ,   Guo et al.  ,
281. **CODE: Contrasting Self-generated Description to Combat Hallucination in Large Multi-modal Models**  ,   Kim et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2406.01920-b31b1b.svg)](https://arxiv.org/abs/2406.01920)

282. **Attention-Space Contrastive Guidance for Efficient Hallucination Mitigation in LVLMs**  ,   Jo et al.  ,
283. **Analyzing and Mitigating Object Hallucination in Large Vision-Language Models**  ,   Zhou et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2310.00754-b31b1b.svg)](https://arxiv.org/abs/2310.00754)

284. **Alleviating Hallucinations in Large Vision-Language Models through Hallucination-Induced Optimization**  ,   Chen et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2405.15356-b31b1b.svg)](https://arxiv.org/abs/2405.15356)

285. **Aligning with Your Own Voice: Self-Corrected Preference Learning for Hallucination Mitigation in LVLMs**  ,   Lim et al.  ,
286. **AHEAD: Attention Head Energy-Aware Dynamics for Hallucination Mitigation in MLLMs**  ,   Chang et al.  ,

## 2023
287. **Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decoding**  ,   Leng et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2311.16922-b31b1b.svg)](https://arxiv.org/abs/2311.16922)

288. **Evaluating Object Hallucination in Large Vision-Language Models**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2305.10355-b31b1b.svg)](https://arxiv.org/abs/2305.10355)

289. **Contrastive Decoding: Open-ended Text Generation as Optimization**  ,   Li et al.  ,   [![arXiv](https://img.shields.io/badge/arXiv-2210.15097-b31b1b.svg)](https://arxiv.org/abs/2210.15097)

awesome-hallucination-mllm

awesome-hallucination-mllm Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

awesome-hallucination-mllm

awesome-hallucination-mllm Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README