humanitas.it

이 이메일 도메인이 관측된 논문 2건

arXiv:2609.10728v1cs.AI

Towards a Deterministic Math Solver for Clinical Language Models

Large language models are unreliable at arithmetic, which is a problem for clinical calculators where a single numerical error changes the recommendation. The standard response is to hardcode each calculator as a validated function, one at a time. We test an alternative: the model does not calculate. Instead, it writes case-specific Python that a restricted local executor runs as a deterministic solver, and the model's task reduces to deciding how to use it. We evaluate this Program-Solve interface on MedCalc-Bench Verified (1,100 cases, 55 calculators) against direct model arithmetic and a hand-written 22-calculator library, using Qwen2.5-7B and Qwen2.5-32B-AWQ, after auditing the benchmark's formulas against current clinical guidelines and flagging 16 of 55 with version, use or coefficient concerns. With formulas and gold variables supplied and both routes reading the whole note, handing off to the solver is not a reliable advantage at 7B (75.31% against 72.02%, a paired +3.29 points with a 95% calculator-cluster interval of [-3.49, 10.38]) but is one at 32B (90.53% against 83.47%, +7.05 [0.47, 14.60], clear of zero). The hand-written library is exact on its 440 supported cases but abstains elsewhere (40.0% overall). Adding an executor thus helps some open-weight models more than others even under matched formula, variable and note access, and is not a substitute for verified formulas or reliable variable extraction either way.

arXiv:2607.06133v1cs.SE

데이터 희소 소프트웨어 시스템을 위한 속성 기반 합성 데이터 엔지니어링: 유방암 도메인에서의 성찰

Property-Driven Synthetic Data Engineering for Data-Scarce Software Systems: Reflections from the Breast Cancer Domain

유방암 수술 중 방사선 치료(IORT) 소프트웨어 개발 경험을 바탕으로, 합성 데이터 생성이 데이터 희소 문제를 해결하는 것이 아니라 핵심 엔지니어링 과제를 이해관계자별 유효성 속성(통계적 충실도, 임상적 타당성, 프라이버시 보호 등) 정의 문제로 전환시킨다는 점을 제시한다. TabDDPM, CTGAN, TVAE 등 다양한 합성 데이터 기법을 실제 IORT 데이터셋에 적용한 예비 실험을 통해 통계적 유사성만으로는 임상적 타당성을 보장할 수 없음을 보이며, 속성 도출·형식화·검증·진화를 위한 자동화 소프트웨어 엔지니어링 연구 방향을 제안한다.