ExperienceWeaver: Optimizing Small-sample Experience Learning for LLM-based Clinical Text Improvement

2026-01-31

AI Summary

This work addresses the challenge of clinical text refinement, which is hindered by the scarcity of high-quality data and the intricate constraints inherent in medical documentation, leading to suboptimal performance of existing large language models in few-shot settings. To overcome this limitation, the authors propose ExperienceWeaver, a novel framework that introduces an experience-based learning mechanism. By identifying error types and distilling multidimensional feedback into error-specific editing tips and high-level revision strategies, ExperienceWeaver enables the model to learn not just what to edit but how to edit through an agent-driven reasoning pipeline. This approach marks a paradigm shift from data retrieval to structured knowledge distillation. Extensive experiments across four clinical datasets demonstrate that ExperienceWeaver significantly outperforms state-of-the-art models such as Gemini-3 Pro, confirming its effectiveness and robustness under few-shot conditions.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

Clinical text improvement is vital for healthcare efficiency but remains difficult due to limited high-quality data and the complex constraints of medical documentation. While Large Language Models (LLMs) show promise, current approaches struggle in small-sample settings: supervised fine-tuning is data-intensive and costly, while retrieval-augmented generation often provides superficial corrections without capturing the reasoning behind revisions. To address these limitations, we propose ExperienceWeaver, a hierarchical framework that shifts the focus from data retrieval to experience learning. Instead of simply recalling past examples, ExperienceWeaver distills noisy, multi-dimensional feedback into structured, actionable knowledge. Specifically, error-specific Tips and high-level Strategies. By injecting this distilled experience into an agentic pipeline, the model learns"how to revise"rather than just"what to revise". Extensive evaluations across four clinical datasets demonstrate that ExperienceWeaver consistently improves performance, surpassing state-of-the-art models such as Gemini-3 Pro in small-sample settings.

Problem

• clinical text improvement
• small-sample learning
• large language models
• medical documentation
• experience learning

Innovation

• Experience Learning
• Small-sample Learning
• Clinical Text Improvement
• Hierarchical Knowledge Distillation
• LLM-based Revision