ASCRIBE: Atomic and Significance-Based Reasoning for Thai Clinical SOAP Note Generation
On the map: LLM reasoning → Clinical and biomedical applications
Abstract (arXiv)
Automatic SOAP note generation can ease the documentation burden on physicians, but existing reasoning methods often omit clinically important information and generate unsupported content. Progress in Thai is further hindered by the lack of publicly available datasets. We propose ASCRIBE, a physician-inspired reasoning framework that ascribes a clinical-significance level to each extracted atomic fact in the conversation before summarization, making a general-purpose LLM a more reliable scribe. We also release ThaiClinicBench, the first de-identified Thai clinical summarization benchmark of real encounters, together with a synthetic training corpus derived from real clinical notes. As a prompt, ASCRIBE outperforms chain-of-thought prompting on GPT-5.4 and Gemini 3.1 Pro across the physician-aligned LLM-judge metrics and improves on standard prompting by up to 10.3 points on the completeness LLM-judge metric. As a GRPO reward, it enables a Gemma-4-E4B model trained solely on synthetic data to match Gemini 3.1 Pro in factual precision and surpass it in completeness. Code and data can be found at https://github.com/loolootech/ascribe.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Clinical and biomedical applications, one of 17 directions in LLM reasoning. Builds on 13 of its 44 references that are in the map. First posted on arXiv 1 Oct 2026.
Builds on
13 of 44 references.
Closest ancestors · 3
Same direction, newest first
- 2026When Reasoning Hurts: Source-Aware Evaluation of Frontier LLMs for Clinical SOAP Note Generation24 May
- 2025Why Chain of Thought Fails in Clinical Text Understanding26 Sep
- 2024Guiding Clinical Reasoning with Large Language Models via Knowledge Seeds11 Mar
Other references · 10
- 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
- 2023FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generationcited by 1,259
- 2020BLEURT: Learning Robust Metrics for Text Generationcited by 1,196
- 2023ACI-BENCH: a Novel Ambient Clinical Intelligence Dataset for Benchmarking Automatic Visit Note Generationcited by 116
- 2020Generating SOAP Notes from Doctor-Patient Conversations Using Modular Summarization Techniquescited by 64
- 2024Exploring Robustness in Doctor-Patient Conversation Summarization: An Analysis of Out-of-Domain SOAP Notescited by 7
- 2024CLINICSUM: Utilizing Language Models for Generating Clinical Summaries from Patient-Doctor Conversationscited by 7
- 2025SERL: Self-Examining Reinforcement Learning on Open-Domaincited by 5
- 2025Assessing Thai Dialect Performance in LLMs with Automatic Benchmarks and Human Evaluationcited by 4
- 2025Optimizing Long-Form Clinical Text Generation with Claim-Based Rewardscited by 3
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025Leaps Beyond the Seen: Reinforced Reasoning Augmented Generation for Clinical Notes
- 2024Process-Supervised Reward Models for Verifying Clinical Note Generation: A Scalable Approach Guided by Domain Expertise
- 2026Note2Chat: Improving LLMs for Multi-Turn Clinical History Taking Using Medical Notes
- 2026AgenticSum: An Agentic Inference-Time Framework for Faithful Clinical Text Summarization
- 2025Enhancing Clinical Note Generation with ICD-10, Clinical Ontology Knowledge Graphs, and Chain-of-Thought Prompting Using GPT-4
- 2022Retrieval-Augmented and Knowledge-Grounded Language Models for Faithful Clinical Medicine
- 2025CLI-RAG: A Retrieval-Augmented Framework for Clinically Structured and Context Aware Text Generation with LLMs
- 2026A rubric landscape for evaluating clinical reasoning in large language models: what exists, what is missing, and what needs to be combined