mlsanityData to 14 Oct 2026

Revealing Perception and Generation Dynamics in LVLMs: Mitigating Hallucinations via Validated Dominance Correction

arXiv 2512.18813 ↗First posted 21 Dec 2025cs.CVcited by 7

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

Abstract (arXiv)

Large Vision-Language Models (LVLMs) have shown remarkable capabilities, yet hallucinations remain a persistent challenge. This work presents a systematic analysis of the internal evolution of visual perception and token generation in LVLMs, revealing two key patterns. First, perception follows a three-stage GATE process: early layers perform a Global scan, intermediate layers Approach and Tighten on core content, and later layers Explore supplementary regions. Second, generation exhibits an SAD (Subdominant Accumulation to Dominant) pattern, where hallucinated tokens arise from the repeated accumulation of subdominant tokens lacking support from attention (visual perception) or feed-forward network (internal knowledge). Guided by these findings, we devise the VDC (Validated Dominance Correction) strategy, which detects unsupported tokens and replaces them with validated dominant ones to improve output reliability. Extensive experiments across multiple models and benchmarks confirm that VDC substantially mitigates hallucinations.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Placed in Hallucination mitigation in LVLMs, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 68 of its 84 references that are in the map. First posted on arXiv 21 Dec 2025.

Builds on

68 of 84 references.

Closest ancestors · 25

Same direction, newest first

  1. 2025ONLY: One-Layer Intervention Sufficiently Mitigates Hallucinations in Large Vision-Language Models1 Jul
  2. 2025SECOND: Mitigating Perceptual Hallucination in Vision-Language Models via Selective and Contrastive Decoding10 Jun
  3. 2025Enhancing Visual Reliance in Text Generation: A Bayesian Perspective on Mitigating Hallucination in Large Vision-Language Models26 May
  4. 2025TARAC: Mitigating Hallucination in LVLMs via Temporal Attention Real-time Accumulative Connection5 Apr
  5. 2025The Hidden Life of Tokens: Reducing Hallucination of Large Vision-Language Models via Visual Information Steering5 Feb
  6. 2024Cracking the Code of Hallucination in LVLMs with Vision-aware Head Divergence18 Dec
  7. 2024Devils in Middle Layers of Large Vision-Language Models: Interpreting, Detecting and Mitigating Object Hallucinations via Attention Lens23 Nov
  8. 2024Seeing Clearly by Layer Two: Enhancing Attention Heads to Alleviate Hallucination in LVLMs15 Nov
  9. 2024MLLM can see? Dynamic Correction Decoding for Hallucination Mitigation15 Oct
  10. 2024Look Twice Before You Answer: Memory-Space Visual Retracing for Hallucination Mitigation in Multimodal Large Language Models4 Oct
  11. 2024Interpreting and Editing Vision-Language Representations to Mitigate Hallucinations3 Oct
  12. 2024Self-Introspective Decoding: Alleviating Hallucinations for Large Vision-Language Models4 Aug
13 more
  1. 2024Paying More Attention to Image: A Training-Free Method for Alleviating Hallucination in LVLMs31 Jul
  2. 2024Mitigating Object Hallucinations in Large Vision-Language Models with Assembly of Global and Local Attention18 Jun
  3. 2024CODE: Contrasting Self-generated Description to Combat Hallucination in Large Multi-modal Models4 Jun
  4. 2024Don't Miss the Forest for the Trees: Attentional Vision Calibration for Large Vision Language Models28 May
  5. 2024Multi-Modal Hallucination Control by Visual Information Grounding20 Mar
  6. 2024HALC: Object Hallucination Reduction via Adaptive Focal-Contrast Decoding1 Mar
  7. 2024A Survey on Hallucination in Large Vision-Language Models1 Feb
  8. 2023OPERA: Alleviating Hallucination in Multi-Modal Large Language Models via Over-Trust Penalty and Retrospection-Allocation29 Nov
  9. 2023Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decoding28 Nov
  10. 2023Woodpecker: Hallucination Correction for Multimodal Large Language Models24 Oct
  11. 2023Analyzing and Mitigating Object Hallucination in Large Vision-Language Models1 Oct
  12. 2023Detecting and Preventing Hallucinations in Large Vision Language Models11 Aug
  13. 2023Evaluating Object Hallucination in Large Vision-Language Models17 May

Other references · 43

  1. 2017Attention Is All You Needcited by 60,361
  2. 2023GPT-4 Technical Reportcited by 18,396
  3. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  4. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  5. 2023Visual Instruction Tuningcited by 7,869
  6. 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
  7. 2023InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuningcited by 2,876
  8. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,711
  9. 2023InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Taskscited by 2,565
  10. 2023Efficient Streaming Language Models with Attention Sinkscited by 2,186
  11. 2022Survey of Hallucination in Natural Language Generationcited by 2,075
  12. 2023MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Modelscited by 1,652
31 more
  1. 2023Video-LLaVA: Learning United Visual Representation by Alignment Before Projectioncited by 1,402
  2. 2020Transformer Feed-Forward Layers Are Key-Value Memoriescited by 1,336
  3. 2024Eyes Wide Shut? Exploring the Visual Shortcomings of Multimodal LLMscited by 735
  4. 2024DeepSeek-VL: Towards Real-World Vision-Language Understandingcited by 733
  5. 2024MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategiescited by 723
  6. 2018Object Hallucination in Image Captioningcited by 669
  7. 2021Knowledge Neurons in Pretrained Transformerscited by 601
  8. 2023mPLUG-Owl2: Revolutionizing Multi-modal Large Language Model with Modality Collaborationcited by 547
  9. 2025Layer by Layer: Uncovering Hidden Representations in Language Modelscited by 311
  10. 2022Confident Adaptive Language Modelingcited by 310
  11. 2023LMDrive: Closed-Loop End-to-End Driving with Large Language Modelscited by 305
  12. 2024Massive Activations in Large Language Modelscited by 294
  13. 2019Depth-Adaptive Transformercited by 236
  14. 2024OmniMedVQA: A New Large-Scale Comprehensive Evaluation Benchmark for Medical LVLMcited by 184
  15. 2024Senna: Bridging Large Vision-Language Models and End-to-End Autonomous Drivingcited by 156
  16. 2025See What You Are Told: Visual Attention Sink in Large Multimodal Modelscited by 132
  17. 2024Knowledge Circuits in Pretrained Transformerscited by 60
  18. 2024Astute RAG: Overcoming Imperfect Retrieval Augmentation and Knowledge Conflicts for Large Language Modelscited by 58
  19. 2025Seeing but Not Believing: Probing the Disconnect Between Visual Attention and Answer Correctness in VLMscited by 55
  20. 2021Incorporating Residual and Normalization Layers into Analysis of Masked Language Modelscited by 51
  21. 2024VisOnlyQA: Large Vision Language Models Still Struggle with Visual Perception of Geometric Informationcited by 39
  22. 2024From Redundancy to Relevance: Information Flow in LVLMs Across Reasoning Taskscited by 38
  23. 2023Neuron-Level Knowledge Attribution in Large Language Modelscited by 37
  24. 2023Analyzing Feed-Forward Blocks in Transformers through the Lens of Attention Mapscited by 33
  25. 2024UniBias: Unveiling and Mitigating LLM Bias through Internal Attention and FFN Manipulationcited by 31
  26. 2024Unveiling the Ignorance of MLLMs: Seeing Clearly, Answering Incorrectlycited by 30
  27. 2025Benchmark Designers Should "Train on the Test Set" to Expose Exploitable Non-Visual Shortcutscited by 29
  28. 2025Adaptive Layer-skipping in Pre-trained LLMscited by 27
  29. 2024STLLaVA-Med: Self-Training Large Language and Vision Assistant for Medical Question-Answeringcited by 21
  30. 2024Memory-Space Visual Prompting for Efficient Vision-Language Fine-Tuningcited by 20
  31. 2025HoneyBee: Data Recipes for Vision-Language Reasonerscited by 13

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 7.

7 papers.

  1. 2026Self-Correcting RAG: Enhancing Faithfulness via MMKP Context Selection and NLI-Guided MCTScited by 5
  2. 2026Revealing and Enhancing Core Visual Regions: Harnessing Internal Attention Dynamics for Hallucination Mitigation in LVLMscited by 4
  3. 2026Towards Interpretable Hallucination Analysis and Mitigation in LVLMs via Contrastive Neuron Steeringcited by 3
  4. 2026TDFNet: Tri-projection Deformable Fusion Network for Panoramic Salient Object Detectioncited by 1
  5. 2026Countering the Over-Reliance Trap: Mitigating Object Hallucination for LVLMs via a Self-Validation Frameworkcited by 0
  6. 2026LightAIR: Lightweight Action Inversion and Riemannian Rectification for Text-based Person Anomaly Searchcited by 0
  7. 2026Beyond Visual Similarity: Entity-Aligned Retrieval for Knowledge-Based Visual Question Answeringcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Generate, but Verify: Reducing Hallucination in Vision-Language Models with Retrospective Resampling
  2. 2026Reducing Object Hallucination in LVLMs via Emphasizing Image-negative Tokens
  3. 2025Not All Tokens and Heads Are Equally Important: Dual-Level Attention Intervention for Hallucination Mitigation
  4. 2024Reducing Hallucinations in Vision-Language Models via Latent Space Steering
  5. 2026Mitigating Hallucinations in Large Vision-Language Models without Performance Degradation
  6. 2024From Pixels to Tokens: Revisiting Object Hallucinations in Large Vision-Language Models
  7. 2026Beyond Visual Enhancement: Adaptive Multi-Context Steering to Mitigate LVLM Hallucinations
  8. 2025Self-Correcting Decoding with Generative Feedback for Mitigating Hallucinations in Large Vision-Language Models