← Question Answering
Papers with Code · CC BY-SA 4.0

OpenBookQA

아카이브 평가 45건 · 마지막 적재/변경

평가 기록

과거 수집 기록이며 최신 순위가 아닙니다. 데이터셋 버전·split·실험 조건과 지표 방향이 검증되지 않은 결과는 직접 비교할 수 없습니다. 날짜는 논문 날짜이며 최신 평가일을 뜻하지 않습니다.

표가 화면보다 넓으면 좌우로 스크롤하여 논문과 구현 링크를 확인하세요.

원본 기록 순서 · 순위 미부여
모델원본 지표·값논문구현
OPT 66B (one-shot)
Accuracy
58.0
BloombergGPT: A Large Language Model for Finance
2023-03-30
코드 1 · 코드 2
PaLM 2-S (1-shot)
Accuracy
57.4
PaLM 2 Technical Report
2023-05-17
코드 1
BiLSTM max-out question-match (WordNet + science fact)
Accuracy
56.3
Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answering
2018-09-08
코드 1
PaLM 2-M (1-shot)
Accuracy
56.2
PaLM 2 Technical Report
2023-05-17
코드 1
BiLSTM max-out question-match (with a science fact)
Accuracy
55.8
Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answering
2018-09-08
코드 1
Bloomberg GPT 50B (1-shot)
Accuracy
51.6
BloombergGPT: A Large Language Model for Finance
2023-03-30
코드 1 · 코드 2
BLOOM 176B (2-shot)
Accuracy
47.2
BloombergGPT: A Large Language Model for Finance
2023-03-30
코드 1 · 코드 2
GPT-NeoX 50B (2-shot)
Accuracy
44.2
BloombergGPT: A Large Language Model for Finance
2023-03-30
코드 1 · 코드 2
LaMini-GPT 1.5B
Accuracy
39.8
LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions
2023-04-27
코드 1
LaMini-T5 738M
Accuracy
36
LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions
2023-04-27
코드 1
LaMini-F-T5 783M
Accuracy
34
LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions
2023-04-27
코드 1
T5-Large 738M
Accuracy
32.8
LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions
2023-04-27
코드 1
GPT-2-XL 1.5B
Accuracy
32
LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions
2023-04-27
코드 1
FLAN-T5-Large 783M
Accuracy
31.2
LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions
2023-04-27
코드 1
Random chance baseline
Accuracy
25
Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answering
2018-09-08
코드 1