mlsanityData to 14 Oct 2026

SDR: Set-Distance Rewards for Radiology Report Generation

arXiv 2606.00440 ↗First posted 30 May 2026cs.AIcited by 3

On the map: LLM reasoning → Clinical and biomedical applications

Abstract (arXiv)

Reinforcement learning with verifiable rewards has rapidly advanced reasoning in vision--language models. However, for chest X-ray report generation, the standard rewards (i.e. exact-match accuracy and step-level processes) are incompatible because the reports consist of unordered and orthogonal findings, rather than a causal reasoning chain. We address this gap with a set-based view: each report is split into sentences and embedded by a frozen sentence transformer, yielding unordered embedding sets. We propose the use of set-to-set distances between generated and reference embeddings as continuous, permutation-invariant rewards. Across two datasets and three vision--language models (Qwen3-VL-2B/4B, Gemma3-4B), post-training with set-to-set distance based rewards via GRPO consistently outperforms supervised fine-tuning and exact-match GRPO on all headline metrics (BERTScore, RadGraph F1 and CheXbert F1 by average %6.80, %7.82 and %4.45 relative improvements respectively). The same set distances also enable test-time best-of-N selection: scoring candidates by their distance to training-report embeddings outperforms random selection on our trained models as well as three closed-source LLMs (Mistral-Small, Gemini-2.5 Flash-Lite, GPT-4o-mini) with on average %16.4 relative improvement on BERTScore. Used as a streaming signal, they support a more efficient form of test-time scaling: pruning low-scoring candidates mid-generation reduces generated tokens by over 50% while preserving the Findings quality of full best-of-N selection. Together these results establish set-distance rewards as a unified signal for both post-training and test-time scaling in chest X-ray report generation. Our code is publicly hrefhttps://anonymous.4open.science/r/Set-Distance-Rewards-CXR-BFDAavailable.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Placed in Clinical and biomedical applications, one of 17 directions in LLM reasoning. Builds on 24 of its 27 references that are in the map. First posted on arXiv 30 May 2026.

Builds on

24 of 27 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025MedReason: Eliciting Factual Medical Reasoning Steps in LLMs via Knowledge Graphs1 Apr
  2. 2025Med-R1: Reinforcement Learning for Generalizable Medical Reasoning in Vision-Language Models18 Mar
  3. 2025MedVLM-R1: Incentivizing Medical Reasoning Capability of Vision-Language Models (VLMs) via Reinforcement Learning26 Feb

Other references · 21

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  3. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  4. 2025Qwen3 Technical Reportcited by 8,820
  5. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  6. 2023Let's Verify Step by Stepcited by 3,789
  7. 2024Improve Mathematical Reasoning in Language Models by Automated Process Supervisioncited by 441
  8. 2025The Lessons of Developing Process Reward Models in Mathematical Reasoningcited by 415
  9. 2023Med-Flamingo: a Multimodal Medical Few-shot Learnercited by 392
  10. 2019Clinically Accurate Chest X-Ray Report Generationcited by 139
  11. 2023Chain of Code: Reasoning with a Language Model-Augmented Code Emulatorcited by 123
  12. 2023Dynamic Graph Enhanced Contrastive Learning for Chest X-ray Report Generationcited by 118
9 more
  1. 2025Process Reward Models That Thinkcited by 118
  2. 2021Multi-modal Understanding and Generation for Medical Images and Text via Vision-Language Pre-Trainingcited by 96
  3. 2024CT2Rep: Automated Radiology Report Generation for 3D Medical Imagingcited by 96
  4. 2024Entropy-Regularized Process Reward Modelcited by 38
  5. 2025R-PRM: Reasoning-Driven Process Reward Modelingcited by 38
  6. 2024Towards a Holistic Framework for Multimodal Large Language Models in Three-dimensional Brain CT Report Generationcited by 31
  7. 2025ReXGradient-160K: A Large-Scale Publicly Available Dataset of Chest Radiographs with Free-text Reportscited by 27
  8. 2021Medical-VLBERT: Medical Visual Language BERT for COVID-19 CT Report Generation With Alternate Learningcited by 24
  9. 2025More Bang for the Buck: Process Reward Modeling with Entropy-Driven Uncertaintycited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026Discrete Diffusion Language Models for Interactive Radiology Report Draftingcited by 2
  2. 2026Transition-Aware best-of-N sampling for Longitudinal Chest X-ray Reportscited by 0
  3. 2026From Sampled Evidence to Radiology Reports: Objectives, Constraints, and Realizationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Enhancing Visual Reasoning in Chest X-Ray Report Generation Using Reinforcement Learning
  2. 2025RadHiera: Semantic Hierarchical Reinforcement Learning for Medical Report Generation
  3. 2023Longitudinal Data and a Semantic Similarity Reward for Chest X-Ray Report Generation
  4. 2026CXRMate-2: Structured Multimodal Temporal Embeddings and Tractable Reinforcement Learning for Clinically Acceptable Chest X-ray Radiology Report Generation
  5. 2025Reason Like a Radiologist: Chain-of-Thought and Reinforcement Learning for Verifiable Report Generation
  6. 2025Enhancing Radiology Report Generation and Visual Grounding using Reinforcement Learning
  7. 2026DobicVLM: Aligning Chest X-Ray Report Generation with Clinically-Grounded Programmatic Rewards via Group Relative Policy Optimization
  8. 2026CARE-X: Towards Clinically Useful Radiology VLMs with Auxiliary Supervision, Reward-Aligned Learning, and Tool-Augmented Measurement