mlsanityData to 7 Oct 2026

Confidence Calibration in Vision-Language-Action Models

arXiv 2507.17383 ↗First posted 23 Jul 2025cs.RO cs.LGcited by 10

On the map: Vision-language-action models → Failure recovery and robustness

Abstract (arXiv)

Trustworthy robot behavior requires not only high levels of task success but also that the robot can reliably quantify how likely it is to succeed. To this end, we present a first-of-its-kind study of confidence calibration in vision-language-action (VLA) foundation models, which map visual observations and natural language instructions to low-level robot motor commands. We establish a confidence baseline for VLAs, examine how task success relates to calibration error and how calibration evolves over time, and introduce two lightweight techniques to remedy the miscalibration we observe: prompt ensembles and action-wise Platt scaling. Our aim in this study is to begin to develop the tools and conceptual understanding necessary to render VLAs both highly performant and highly trustworthy via reliable uncertainty quantification.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Placed in Failure recovery and robustness, one of 13 directions in Vision-language-action models. Builds on 34 of its 47 references that are in the map. First posted on arXiv 23 Jul 2025.

Builds on

34 of 47 references.

References · 34

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2017On Calibration of Modern Neural Networkscited by 4,860
  3. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  4. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,888
  5. 2015Continuous control with deep reinforcement learningcited by 3,279
  6. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 2,993
  7. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,777
  8. 2024π0: A Vision-Language-Action Flow Model for General Robot Controlcited by 2,297
  9. 2022Language Models (Mostly) Know What They Knowcited by 1,783
  10. 2025π0.5: a Vision-Language-Action Model with Open-World Generalizationcited by 1,653
  11. 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,338
  12. 2019Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shiftcited by 1,274
22 more
  1. 2023Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generationcited by 795
  2. 2023Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedbackcited by 788
  3. 2018Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Modelscited by 748
  4. 2022Teaching Models to Express Their Uncertainty in Wordscited by 731
  5. 2021Revisiting the Calibration of Modern Neural Networkscited by 387
  6. 2025UniVLA: Learning to Act Anywhere with Task-centric Latent Actionscited by 386
  7. 2020Calibrating Deep Neural Networks using Focal Losscited by 353
  8. 2024INSIDE: LLMs' Internal States Retain the Power of Hallucination Detectioncited by 317
  9. 2023Analyzing and Mitigating Object Hallucination in Large Vision-Language Modelscited by 314
  10. 2019Verified Uncertainty Calibrationcited by 307
  11. 2023Robots That Ask For Help: Uncertainty Alignment for Large Language Model Plannerscited by 298
  12. 2023Generating with Confidence: Uncertainty Quantification for Black-box Large Language Modelscited by 286
  13. 2019Self-Supervised Exploration via Disagreementcited by 263
  14. 2017Cost-Effective Active Learning for Deep Image Classificationcited by 165
  15. 2023Shifting Attention to Relevance: Towards the Predictive Uncertainty Quantification of Free-Form Large Language Modelscited by 147
  16. 2025NORA: A Small Open-Sourced Generalist Vision Language Action Model for Embodied Taskscited by 119
  17. 2024Linguistic Calibration of Long-Form Generationscited by 59
  18. 2022Calibrated Selective Classificationcited by 30
  19. 2023COPlanner: Plan to Roll Out Conservatively but to Explore Optimistically for Model-Based RLcited by 14
  20. 2024Ensemble everything everywhere: Multi-scale aggregation for adversarial robustnesscited by 12
  21. 2024DADEE: Well-calibrated uncertainty quantification in neural networks for barriers-based robot safetycited by 3
  22. 2024Improving Predictor Reliability with Selective Recalibrationcited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 9.

10 papers.

  1. 2026Vision-Language-Action Safety: Threats, Challenges, Evaluations, and Mechanismscited by 12
  2. 2026ReconVLA: An Uncertainty-Guided and Failure-Aware Vision-Language-Action Framework for Robotic Controlcited by 5
  3. 2026When to Act, Ask, or Learn: Uncertainty-Aware Policy Steeringcited by 4
  4. 2026SCALE: Self-uncertainty Conditioned Adaptive Looking and Execution for Vision-Language-Action Modelscited by 3
  5. 2025Prune-Then-Plan: Step-Level Calibration for Stable Frontier Exploration in Embodied Question Answeringcited by 2
  6. 2026Calibrated Triage, Not Autonomy: Confidence Estimation for Medical Vision-Language Modelscited by 2
  7. 2026Temporal Difference Calibration in Sequential Tasks: Application to Vision-Language-Action Modelscited by 1
  8. 2026VLAConf: Calibrated Task-Success Confidence for Vision-Language-Action Modelscited by 1
  9. 2026Confidence Estimation for Financial Vision-Language Models in Chart and Document Understandingcited by 0
  10. 2026FabriMAE I Trust Myself? Self-Evaluating VLA Action Generation with Markov Attention Entropycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Evaluating Uncertainty and Quality of Vision-Language-Action-enabled Robots
  2. 2025Seeing is Believing, but How Much? A Comprehensive Analysis of Verbalized Calibration in Vision-Language Models
  3. 2025VLA-Arena: An Open-Source Framework for Benchmarking Vision-Language-Action Models
  4. 2026Is Success All You Need? Investigating the Impact of Input Perturbations on VLA Behaviour in Tabletop Manipulation Tasks
  5. 2025LIBERO-Plus: In-depth Robustness Analysis of Vision-Language-Action Models
  6. 2026Perturbation-Based Epistemic Uncertainty for Failure Detection in Vision-Language-Action Models
  7. 2026Shifting Uncertainty to Critical Moments: Towards Reliable Uncertainty Quantification for VLA Model
  8. 2026Rethinking Visual-Language-Action Model Scaling: Alignment, Mixture, and Regularization