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

Knowledge Integration Quality on NQ, TruthfulQA, WoW, HotpotQA, ELI5 Aggregate

76.7Average Performance

GuarantRAG

Updated 1mo ago

Evaluation Results

Method	Links
GuarantRAG 2026.04		76.7	18.4	89.3	4.74
o1 2026.04		76.4	12.8	-	4.85
GuarantRAG 2026.04		75.8	13.2	90.9	4.72
GPT-4o 2026.04		75	14.1	-	4.81
GuarantRAG 2026.04		74	15.7	88.4	4.68
P-RAG 2026.04		73.8	25.6	86.9	4.49
P-RAG 2026.04		72.5	20.4	88.7	4.47
SOLAR 2026.04		71.9	27.1	85.1	4.43
DRAGIN 2026.04		71.5	28.2	84.7	4.44
FLARE 2026.04		71.1	28.7	84.3	4.41
SOLAR 2026.04		70.5	22.3	86.8	4.41
RQ-RAG 2026.04		70.3	29.3	83.4	4.37
DRAGIN 2026.04		70	23	86.5	4.42
FLARE 2026.04		69.6	23.5	86.1	4.39
P-RAG 2026.04		69	22.9	86.1	4.43
RQ-RAG 2026.04		68.9	24.1	85.2	4.35
SelfRAG 2026.04		68.6	31.2	81.9	4.34
SOLAR 2026.04		67.5	24.7	84.5	4.37
DA-RAG 2026.04		67.5	32.5	80.7	4.28
SelfRAG 2026.04		67.2	25.9	83.7	4.32
DRAGIN 2026.04		67.1	25.3	84.2	4.38
FLARE 2026.04		66.7	25.8	83.7	4.35
DA-RAG 2026.04		66.2	27.2	82.4	4.26
RQ-RAG 2026.04		65.9	26.4	82.9	4.31
SelfRAG 2026.04		64.4	28.1	81.3	4.28
Standard RAG 2026.04		64	34.7	77.2	4.21
DA-RAG 2026.04		63.3	29.5	80.1	4.22
Standard RAG 2026.04		62.8	30.4	78.9	4.19
Standard RAG 2026.04		60	32.6	76.8	4.15
Qwen3-14B-T 2026.04		59.5	32.9	-	4.79
Qwen3-14B 2026.04		57.3	35.8	-	4.76
Qwen3-8B 2026.04		54.8	38.2	-	4.72