mlsanityData to 14 Oct 2026

When Relations Break: Analyzing Relation Hallucination in Vision-Language Model Under Rotation and Noise

arXiv 2605.05045 ↗First posted 6 May 2026cs.CV cs.CLcited by 0

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

Abstract (arXiv)

Vision-language models (VLMs) achieve strong multimodal performance but remain prone to relation hallucination, which requires accurate reasoning over inter-object interactions. We study the impact of visual perturbations, specifically rotation and noise, and show that even mild distortions significantly degrade relational reasoning across models and datasets. We further evaluate prompt-based augmentation and preprocessing strategies (orientation correction and denoising), finding that while they offer partial improvements, they do not fully resolve hallucinations. Our results reveal a gap between perceptual robustness and relational understanding, highlighting the need for more robust, geometry-aware VLMs.

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 14 of its 24 references that are in the map. First posted on arXiv 6 May 2026.

Builds on

14 of 24 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024Reefknot: A Comprehensive Benchmark for Relation Hallucination Evaluation, Analysis and Mitigation in Multimodal Large Language Models18 Aug
  2. 2024Evaluating and Analyzing Relationship Hallucinations in Large Vision-Language Models24 Jun

Other references · 12

  1. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  2. 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
  3. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,234
  4. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,877
  5. 2023InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Taskscited by 2,565
  6. 2024Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wildcited by 286
  7. 2025VLM-R3: Region Recognition, Reasoning, and Refinement for Enhanced Multimodal Chain-of-Thoughtcited by 49
  8. 2025UniRestore: Unified Perceptual and Task-Oriented Image Restoration Model Using Diffusion Priorcited by 42
  9. 2024LLMs Meet VLMs: Boost Open Vocabulary Object Detection with Fine-grained Descriptorscited by 40
  10. 2024B-AVIBench: Towards Evaluating the Robustness of Large Vision-Language Model on Black-box Adversarial Visual-Instructionscited by 26
  11. 2025Losing the Plot: How VLM responses degrade on imperfect chartscited by 5
  12. 2025RotBench: Evaluating Multimodal Large Language Models on Identifying Image Rotationcited by 4

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Mitigating Hallucinations in Multimodal Spatial Relations through Constraint-Aware Prompting
  2. 2025ChainMPQ: Interleaved Text-Image Reasoning Chains for Mitigating Relation Hallucinations
  3. 2026RECAP: Relation Evidence Calibration for Detecting Spatial Relation Hallucinations in Vision-Language Models
  4. 2026HIVE: Understanding Post-Hallucination Reasoning in Vision Language Models
  5. 2026Mitigating Action-Relation Hallucinations in LVLMs via Relation-aware Visual Enhancement
  6. 2025SegSub: Evaluating Robustness to Knowledge Conflicts and Hallucinations in Vision-Language Models
  7. 2026RelCheck: Dual-Evidence Spatial Grounding for VLM Hallucination Correction
  8. 2026A Progressive Training Strategy for Embodied Vision-Language Models to Mitigate Spatio-Temporal Hallucinations