독서리서치레이더투자 프레임워크
로그인 / 회원가입
← 읽기로 돌아가기
뉴스간단한 요점1 분 소요DisconfirmAI 편집부

Surgical video QA benchmark self-reports 14,256 pairs and 14.61% gain

SurgRAW authors self-report SurgCoTBench with 14,256 QA pairs and retrieval-augmented multi-agent reasoning beating supervised models by 14.61%, with no independent verification yet.

검증됨 2026-09-18 00:59 GMT+8

Surgical video AI now has a question-answering benchmark covering five surgical task types, SurgCoTBench, which the authors self-report contains 14,256 QA pairs, with retrieval-augmented multi-agent reasoning accuracy exceeding supervised models by 14.61%.

Previously the field lacked a unified surgical video question-answering benchmark, making it hard to compare methods across five surgical tasks.

Chang Han Low et al. self-reported the above benchmark and the 14.61% accuracy gain, updating arXiv v3 on September 16, with the entry marked as citing IEEE RA-L 2026; code is open on GitHub and the dataset has been released.

No independent verification yet.

Sources:

  • arXiv abstract page (v3, revised 2026-09-16) ↗
  • SurgRAW code repository (GitHub) ↗

출처

  • arxiv.org↗
  • github.com↗
로그인 / 회원가입
독서리서치레이더투자 프레임워크
ieeexplore.ieee.org
↗
리서치 계속하기 →