mlsanityData to 14 Oct 2026

Leveraging Visual Signals for Robust Token-Level Uncertainty in Vision-Language Generation

arXiv 2605.27136 ↗First posted 26 May 2026cs.CVcited by 0

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

Abstract (arXiv)

Uncertainty quantification (UQ) remains a critical challenge in Large Vision Language Models (LVLMs) for reliable predictions and real-world deployment. However, most existing methods are adapted from the LLM literature and primarily focus on the language modality, leaving the contribution of visual information to LVLM uncertainty largely underexplored. In this paper, we investigate how LVLMs process visual information and whether this process can be used to improve uncertainty estimation. By analyzing hidden representations after the integration of visual features during the generation process, we observe that high-confidence predictions rely more heavily on visual content than uncertain ones. Building on this insight, we propose Visual-Grounded Token UQ (VIG-TUQ), a training-free framework that explicitly incorporates visual grounding into uncertainty estimation by weighting token-level language uncertainty with visual grounding scores. We evaluate VIG-TUQ on multiple datasets and across diverse LVLM architectures, including early-fusion, late-fusion, and native-fusion models. Results indicate that our method often improves upon existing token-level uncertainty approaches. Code and data will be made available upon acceptance.

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 106 of its 133 references that are in the map. First posted on arXiv 26 May 2026.

Builds on

106 of 133 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025Improving Semantic Uncertainty Quantification in LVLMs with Semantic Gaussian Processes16 Dec
  2. 2024Post-hoc Probabilistic Vision-Language Models8 Dec
  3. 2024VL-Uncertainty: Detecting Hallucination in Large Vision-Language Model via Uncertainty Estimation18 Nov
  4. 2024MetaToken: Detecting Hallucination in Image Descriptions by Meta Classification29 May
  5. 2024A Survey on Hallucination in Large Vision-Language Models1 Feb

Other references · 101

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  2. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  3. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  4. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  5. 2023Visual Instruction Tuningcited by 7,869
  6. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  7. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,137
  8. 2025Qwen2.5-VL Technical Reportcited by 5,587
  9. 2017On Calibration of Modern Neural Networkscited by 4,960
  10. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  11. 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
  12. 2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,386
38 more
  1. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,926
  2. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,955
  3. 2023InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuningcited by 2,876
  4. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,711
  5. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,385
  6. 2023Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyondcited by 2,224
  7. 2020DeBERTa: Decoding-enhanced BERT with Disentangled Attentioncited by 2,181
  8. 2022Survey of Hallucination in Natural Language Generationcited by 2,075
  9. 2022Language Models (Mostly) Know What They Knowcited by 1,856
  10. 2023A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questionscited by 1,601
  11. 2023Inference-Time Intervention: Eliciting Truthful Answers from a Language Modelcited by 1,302
  12. 2019OK-VQA: A Visual Question Answering Benchmark Requiring External Knowledgecited by 1,121
  13. 2023Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMscited by 974
  14. 2025AlphaEvolve: A coding agent for scientific and algorithmic discoverycited by 867
  15. 2018VizWiz Grand Challenge: Answering Visual Questions from Blind Peoplecited by 858
  16. 2020A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challengescited by 819
  17. 2023Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generationcited by 819
  18. 2023SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Modelscited by 810
  19. 2022Teaching Models to Express Their Uncertainty in Wordscited by 751
  20. 2023The Internal State of an LLM Knows When It's Lyingcited by 659
  21. 2018Loss Surfaces, Mode Connectivity, and Fast Ensembling of DNNscited by 588
  22. 2019A Simple Baseline for Bayesian Uncertainty in Deep Learningcited by 560
  23. 2024BLINK: Multimodal Large Language Models Can See but Not Perceivecited by 550
  24. 2020Bayesian Deep Learning and a Probabilistic Perspective of Generalizationcited by 480
  25. 2019Deep Ensembles: A Loss Landscape Perspectivecited by 443
  26. 2020Simple and Principled Uncertainty Estimation with Deterministic Deep Learning via Distance Awarenesscited by 389
  27. 2020BatchEnsemble: An Alternative Approach to Efficient Ensemble and Lifelong Learningcited by 371
  28. 2021Laplace Redux -- Effortless Bayesian Deep Learningcited by 365
  29. 2024Hallucination is Inevitable: An Innate Limitation of Large Language Modelscited by 332
  30. 2024What matters when building vision-language models?cited by 326
  31. 2024INSIDE: LLMs' Internal States Retain the Power of Hallucination Detectioncited by 323
  32. 2023OBELICS: An Open Web-Scale Filtered Dataset of Interleaved Image-Text Documentscited by 316
  33. 2025Layer by Layer: Uncovering Hidden Representations in Language Modelscited by 311
  34. 2021What Are Bayesian Neural Network Posteriors Really Like?cited by 308
  35. 2015Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networkscited by 295
  36. 2023Generating with Confidence: Uncertainty Quantification for Black-box Large Language Modelscited by 293
  37. 2021Ensemble deep learning: A reviewcited by 284
  38. 2024Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMscited by 259

50 most cited of 101 All 101 (JSON)

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026VLM-UQBench: A Benchmark for Modality-Specific and Cross-Modality Uncertainties in Vision Language Models
  2. 2024Uncertainty-Aware Evaluation for Vision-Language Models
  3. 2026VAUQ: Vision-Aware Uncertainty Quantification for LVLM Self-Evaluation
  4. 2025Exploiting the Asymmetric Uncertainty Structure of Pre-trained VLMs on the Unit Hypersphere
  5. 2025The Art of Saying "Maybe": A Conformal Lens for Uncertainty Benchmarking in VLMs
  6. 2026FUSE: Quantifying Uncertainty in Vision-Language Models by Bayesian Fusing Epistemic and Aleatoric Uncertainty
  7. 2026Dynamic Distribution-Aware Uncertainty Tracking in Vision-Language Representation Learning
  8. 2025ViLU: Learning Vision-Language Uncertainties for Failure Prediction