Optimizing Long-Form Clinical Text Generation with Claim-Based Rewards
This work addresses the challenge of generating accurate and complete clinical notes for healthcare professionals, representing an incremental improvement over existing methods.
The authors tackled the problem of automating clinical documentation by developing a reinforcement learning framework that optimizes long-form clinical text generation for factual grounding and completeness, resulting in improved note quality and reduced training costs with higher preference for outputs in factuality, completeness, and brevity.
Automating clinical documentation with large language models requires precise alignment with priorities such as completeness and factual grounding. We present an evaluation-integrated reinforcement learning framework for long-form clinical text generation that couples Group Relative Policy Optimization (GRPO) with DocLens, a claim-level evaluator that provides deterministic, dialogue-grounded rewards. Our method directly optimizes factual grounding and completeness without training a separate reward model or relying on human-authored references. Empirically, the approach improves clinical note quality and reduces training cost via a simple reward-gating strategy. An independent GPT-5 qualitative evaluation further supports these gains, showing higher preference for GRPO outputs in factuality, completeness, and brevity, with fewer omissions and hallucinations. Because the benchmarks are relatively clean and the base model already well aligned, these improvements likely represent a conservative lower bound. The framework is scalable to real-world settings and can incorporate custom objectives such as guideline adherence or billing preferences.