Share your thoughts, 1 month free Claude Pro on usSee more

Open-Domain Question Answering on NaturalQuestions (NQ)

52.88SubEM

Stable-RAG

Updated 1mo ago

Evaluation Results

Method	Links
Stable-RAG 2026.01		52.88	51.78
Stable-RAG 2026.01		52.12	52.38
Stable-RAG 2026.01		52.02	50.72
Stable-RAG 2026.01		51.69	51.78
ATM 2026.01		50.94	51.03
Pos2Distill 2026.01		50.71	50.93
Vanilla RAG 2026.01		50.55	50.67
RAAT 2026.01		50.12	50.03
Ms-PoE 2026.01		50.04	50.08
ATM 2026.01		49.78	50.19
RetRobust 2026.01		49.5	50.81
Pos2Distill 2026.01		49.25	48.37
RAAT 2026.01		49.12	49.85
Stable-RAG 2026.01		48.75	46.58
RetRobust 2026.01		48.7	49.29
Stable-RAG 2026.01		48.14	45.8
Stable-RAG 2026.01		46.94	47.13
Vanilla SFT 2026.01		46.2	49.44
Stable-RAG 2026.01		46.12	46.79
Vanilla RAG 2026.01		45.81	47.8
Vanilla SFT 2026.01		45.6	49.19
Ms-PoE 2026.01		45.58	47.53
ATM 2026.01		45.47	45.86
RAAT 2026.01		45.13	45.87
Pos2Distill 2026.01		44.89	45.52
Vanilla RAG 2026.01		44.65	45.34
Pos2Distill 2026.01		44.58	43.12
Ms-PoE 2026.01		44.39	45.12
ATM 2026.01		43.75	44.88
RetRobust 2026.01		43.1	44.99
RAAT 2026.01		42.33	43.85
Vanilla SFT 2026.01		42.1	44.78
RetRobust 2026.01		41.82	44.26
Vanilla SFT 2026.01		41.41	45.05
Vanilla RAG 2026.01		40.75	42.82
Ms-PoE 2026.01		40.32	42.49
Direct Generation 2026.01		25.18	29.11
Direct Generation 2026.01		25.18	29.11
Naive Generation 2026.01		21.94	24.07
Naive Generation 2026.01		21.94	24.07