Non-reasoning GPT-5.4 outperforms its reasoning version on SOAP note quality across datasets, with RAG giving smaller model-dependent gains.
arXiv preprint arXiv:2507.06715
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
When Reasoning Hurts: Source-Aware Evaluation of Frontier LLMs for Clinical SOAP Note Generation
Non-reasoning GPT-5.4 outperforms its reasoning version on SOAP note quality across datasets, with RAG giving smaller model-dependent gains.