mlsanityData to 14 Oct 2026

Seeing is Believing, but How Much? A Comprehensive Analysis of Verbalized Calibration in Vision-Language Models

arXiv 2505.20236 ↗First posted 26 May 2025cs.CVcited by 16

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

Abstract (arXiv)

Uncertainty quantification is essential for assessing the reliability and trustworthiness of modern AI systems. Among existing approaches, verbalized uncertainty, where models express their confidence through natural language, has emerged as a lightweight and interpretable solution in large language models (LLMs). However, its effectiveness in vision-language models (VLMs) remains insufficiently studied. In this work, we conduct a comprehensive evaluation of verbalized confidence in VLMs, spanning three model categories, four task domains, and three evaluation scenarios. Our results show that current VLMs often display notable miscalibration across diverse tasks and settings. Notably, visual reasoning models (i.e., thinking with images) consistently exhibit better calibration, suggesting that modality-specific reasoning is critical for reliable uncertainty estimation. To further address calibration challenges, we introduce Visual Confidence-Aware Prompting, a two-stage prompting strategy that improves confidence alignment in multimodal settings. Overall, our study highlights the inherent miscalibration in VLMs across modalities. More broadly, our findings underscore the fundamental importance of modality alignment and model faithfulness in advancing reliable multimodal systems.

Key facts. Cited by 16 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 27 of its 52 references that are in the map. First posted on arXiv 26 May 2025.

Builds on

27 of 52 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025Object-Level Verbalized Confidence Calibration in Vision-Language Models via Semantic Perturbation21 Apr
  2. 2025Know What You do Not Know: Verbalized Uncertainty Estimation Robustness on Corrupted Images in Vision-Language Models4 Apr

Other references · 25

  1. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  2. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,719
  3. 2025InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Modelscited by 1,834
  4. 2023MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contextscited by 1,614
  5. 2023Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMscited by 974
  6. 2023Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedbackcited by 812
  7. 2022Teaching Models to Express Their Uncertainty in Wordscited by 751
  8. 2024Measuring Multimodal Mathematical Reasoning with MATH-Vision Datasetcited by 526
  9. 2024MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmarkcited by 512
  10. 2025Video-MMMU: Evaluating Knowledge Acquisition from Multi-Discipline Professional Videoscited by 236
  11. 2024VoiceBench: Benchmarking LLM-Based Voice Assistantscited by 200
  12. 2024Fact-Checking the Output of Large Language Models via Token-Level Uncertainty Quantificationcited by 162
13 more
  1. 2024Kernel Language Entropy: Fine-grained Uncertainty Quantification for LLMs from Semantic Similaritiescited by 159
  2. 2023LM-Polygraph: Uncertainty Estimation for Language Modelscited by 131
  3. 2024SaySelf: Teaching LLMs to Express Confidence with Self-Reflective Rationalescited by 105
  4. 2024Benchmarking Uncertainty Quantification Methods for Large Language Models with LM-Polygraphcited by 105
  5. 2024On Verbalized Confidence Scores for LLMscited by 83
  6. 2025Skywork R1V: Pioneering Multimodal Reasoning with Chain-of-Thoughtcited by 69
  7. 2025Cross the Gap: Exposing the Intra-modal Misalignment in CLIP via Modality Inversioncited by 43
  8. 2024IsoBench: Benchmarking Multimodal Foundation Models on Isomorphic Representationscited by 38
  9. 2025Large Vision-Language Model Alignment and Misalignment: A Survey Through the Lens of Explainabilitycited by 28
  10. 2023Do Vision and Language Models Share Concepts? A Vector Space Alignment Studycited by 27
  11. 2024Unveiling Uncertainty: A Deep Dive into Calibration and Performance of Multimodal Large Language Modelscited by 19
  12. 2025Uncertainty Distillation: Teaching Language Models to Express Semantic Confidencecited by 13
  13. 2023Text as Images: Can Multimodal Large Language Models Follow Printed Instructions in Pixels?cited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 14.

16 papers.

Sort:

By area:

  1. 2025Roboflow100-VL: A Multi-Domain Object Detection Benchmark for Vision-Language Modelscited by 20
  2. 2026The Confidence Dichotomy: Analyzing and Mitigating Miscalibration in Tool-Use Agentscited by 16
  3. 2025Anthropomimetic Uncertainty: What Verbalized Uncertainty in Language Models is Missingcited by 14
  4. 2026Overconfidence and Calibration in Medical VQA: Empirical Findings and Hallucination-Aware Mitigationcited by 6
  5. 2026Instinct vs. Reflection: Unifying Token and Verbalized Confidence in Multimodal Large Modelscited by 4
  6. 2026Just how sure are you? Improving Verbalized Uncertainty Calibration in Medical VQAcited by 3
  7. 2026Seeing Before Agreeing: Aligning Multi-Agent Consensus with Visual Evidencecited by 1
  8. 2026ReLope: KL-Regularized LoRA Probes for Multimodal LLM Routingcited by 0
  9. 2026Uncertainty Quantification for Multimodal Retrieval Augmented Generationcited by 0
  10. 2026VLM-GLoc: Vision-Language Model Enhanced Monte Carlo Localization for Robust Semantic Global Localization in Cluttered Quasi-Static Environmentscited by 0
  11. 2026Bayesian Adaptation Gym: A Benchmark for the Bayesian Low-Rank Adaptation of Multi-Modal Language Modelscited by 0
  12. 2026Visual Semantic Entropy: Do Vision Language Models Recognize Visual Ambiguity?cited by 0
4 more
  1. 2026Evaluation Pitfalls and Sparsity Limitations in LLM-based Confidence Estimates for Classificationcited by 0
  2. 2026Uncertainty-Aware Decision Making in Multimodal Large Language Modelscited by 0
  3. 2026Dissecting Training-Free Uncertainty Estimation in Multimodal Large Language Modelscited by 0
  4. 2026Beyond Saying Less: Fine-Grained Alignment for Informative and Faithful Vision-Language Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026VL-Calibration: Decoupled Confidence Calibration for Large Vision-Language Models Reasoning
  2. 2026The Mirage of Calibrated Confidence: Trajectory-Independence of Verbalized Confidence in Vision-Language Models
  3. 2025The Art of Saying "Maybe": A Conformal Lens for Uncertainty Benchmarking in VLMs
  4. 2024Overconfidence is Key: Verbalized Uncertainty Evaluation in Large Language and Vision-Language Models
  5. 2025The Role of Model Confidence on Bias Effects in Measured Uncertainties for Vision-Language Models
  6. 2025Confidence Calibration in Vision-Language-Action Models
  7. 2026VLM-UQBench: A Benchmark for Modality-Specific and Cross-Modality Uncertainties in Vision Language Models
  8. 2024Uncertainty-Aware Evaluation for Vision-Language Models