← Question Answering

Papers with Code · CC BY-SA 4.0

TrecQA

아카이브 평가 13건 · 마지막 적재/변경 2026-10-30

원본 데이터셋 링크 1

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
모델	원본 지표·값	논문	구현
TANDA DeBERTa-V3-Large + ALL	MAP 0.954 MRR 0.984	Structural Self-Supervised Objectives for Transformers 2023-09-15	코드 1
TANDA-RoBERTa (ASNQ, TREC-QA)	MAP 0.943 MRR 0.974	TANDA: Transfer and Adapt Pre-Trained Transformer Models for Answer Sentence Selection 2019-11-11	코드 1 · 코드 2
DeBERTa-V3-Large + SSP	MAP 0.923 MRR 0.946	Pre-training Transformer Models with Sentence-Level Objectives for Answer Sentence Selection 2022-05-20	연결 없음
Contextual DeBERTa-V3-Large + SSP	MAP 0.919 MRR 0.945	Context-Aware Transformer Pre-Training for Answer Sentence Selection 2023-05-24	연결 없음
RLAS-BIABC	MAP 0.913 MRR 0.998	RLAS-BIABC: A Reinforcement Learning-Based Answer Selection Using the BERT Model Boosted by an Improved ABC Algorithm 2023-01-07	연결 없음
RoBERTa-Base Joint + MSPP	MAP 0.911 MRR 0.952	Paragraph-based Transformer Pre-training for Multi-Sentence Inference 2022-05-02	코드 1
RoBERTa-Base + PSD	MAP 0.903 MRR 0.951	Pre-training Transformer Models with Sentence-Level Objectives for Answer Sentence Selection 2022-05-20	연결 없음
Comp-Clip + LM + LC	MAP 0.868 MRR 0.928	A Compare-Aggregate Model with Latent Clustering for Answer Selection 2019-05-30	연결 없음
NLP-Capsule	MAP 0.7773 MRR 0.7416	Towards Scalable and Reliable Capsule Networks for Challenging NLP Applications 2019-06-06	코드 1 · 코드 2 · 코드 3 · 코드 4 · 코드 5
HyperQA	MAP 0.770 MRR 0.825	Hyperbolic Representation Learning for Fast and Efficient Neural Question Answering 2017-07-25	코드 1
PWIN	MAP 0.7588 MRR 0.8219	Pairwise Word Interaction Modeling with Deep Neural Networks for Semantic Similarity Measurement 2016-06-01	연결 없음
aNMM	MAP 0.750 MRR 0.811	aNMM: Ranking Short Answer Texts with Attention-Based Neural Matching Model 2018-01-05	코드 1
CNN	MAP 0.711 MRR 0.785	Deep Learning for Answer Sentence Selection 2014-12-04	코드 1 · 코드 2