mlsanityData to 13 Oct 2026

CARE: Confidence-Aware Reasoning for Reliable Medical VQA

arXiv 2608.10964 ↗First posted 11 Aug 2026cs.CV cs.AIcited by 0

On the map: LLM reasoning → Clinical and biomedical applications

also in 1 more
Abstract (arXiv)

Reinforcement Fine-Tuning (RFT) has enabled medical Multimodal Large Language Models (MLLMs) to produce Chain-of-Thought (CoT) reasoning for visual question answering, yet these models suffer from textitconfidence miscalibration---a systematic gap between expressed certainty and actual diagnostic accuracy that undermines clinical trust. We propose CARE, a Confidence-Aware medical REasoning framework that jointly optimizes accuracy and calibration through a dual-stage pipeline. First, a scalable Medical-CoT synthesis provides structured cold-start data for Supervised Fine-Tuning. Second, Group Relative Policy Optimization (GRPO) with a novel Confidence-Aware Reward (CAR) mechanism ties the model's confidence to diagnostic correctness within the reward signal. Across three Medical VQA benchmarks, CARE achieves the highest diagnostic accuracy while obtaining the lowest Expected Calibration Error and Hallucination Rate, establishing a foundation for trustworthy clinical decision support. Our code is available at https://github.com/anotherbricki/CARE.

Key facts. No citing papers in the map yet (data through 13 Oct 2026). Placed in Clinical and biomedical applications, one of 17 directions in LLM reasoning. Builds on 19 of its 22 references that are in the map. First posted on arXiv 11 Aug 2026.

Builds on

19 of 22 references.

Closest ancestors · 6

Same direction, newest first

  1. 2025Fleming-VL: Towards Universal Medical Visual Reasoning with Multimodal LLMs2 Nov
  2. 2025MedVLThinker: Simple Baselines for Multimodal Medical Reasoning4 Aug
  3. 2025Chiron-o1: Igniting Multimodal Large Language Models towards Generalizable Medical Reasoning via Mentor-Intern Collaborative Search20 Jun
  4. 2025Lingshu: A Generalist Foundation Model for Unified Multimodal Medical Understanding and Reasoning8 Jun
  5. 2025Med-R1: Reinforcement Learning for Generalizable Medical Reasoning in Vision-Language Models18 Mar
  6. 2025MedVLM-R1: Incentivizing Medical Reasoning Capability of Vision-Language Models (VLMs) via Reinforcement Learning26 Feb

Other references · 13

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,091
  2. 2017On Calibration of Modern Neural Networkscited by 4,902
  3. 2023LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Daycited by 1,296
  4. 2024Tulu 3: Pushing Frontiers in Open Language Model Post-Trainingcited by 1,020
  5. 2023Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMscited by 960
  6. 2023Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedbackcited by 800
  7. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 663
  8. 2020PathVQA: 30000+ Questions for Medical Visual Question Answeringcited by 472
  9. 2021SLAKE: A Semantically-Labeled Knowledge-Enhanced Dataset for Medical Visual Question Answeringcited by 446
  10. 2025From System 1 to System 2: A Survey of Reasoning Large Language Modelscited by 320
  11. 2025Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Modelscited by 292
  12. 2026MedMO: Grounding and Understanding Multimodal Large Language Model for Medical Imagescited by 15
1 more
  1. 2026Confidence Calibration for Multimodal LLMs: An Empirical Study through Medical VQAcited by 4

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Adaptive Reinforcement for Open-ended Medical Reasoning via Semantic-Guided Reward Collapse Mitigation
  2. 2025Toward Effective Reinforcement Learning Fine-Tuning for Medical VQA in Vision-Language Models
  3. 2026Overconfidence and Calibration in Medical VQA: Empirical Findings and Hallucination-Aware Mitigation
  4. 2026Breaking Failure Cascades: Step-Aware Reinforcement Learning for Medical Multimodal Reasoning
  5. 2025AdaThink-Med: Optimizing Inference-Time Compute for Medical Reasoning via Uncertainty Quantification
  6. 2025Improving Medical Reasoning with Curriculum-Aware Reinforcement Learning
  7. 2025Prompt4Trust: A Reinforcement Learning Prompt Augmentation Framework for Clinically-Aligned Confidence Calibration in Multimodal Large Language Models
  8. 2025MedMMV: A Controllable Multimodal Multi-Agent Framework for Reliable and Verifiable Clinical Reasoning