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

Multi-hop Question Answering on HotpotQA (Performance Score)

78.2Performance Score

GuarantRAG

Updated 8d ago

Evaluation Results

Method	Links
GuarantRAG 2026.04		78.2
P-RAG 2026.04		74.9
SOLAR 2026.04		72.5
o1 2026.04		72.3
DRAGIN 2026.04		72.1
FLARE 2026.04		71.7
GuarantRAG 2026.04		71.6
GPT-4o 2026.04		70.9
RQ-RAG 2026.04		70.8
SelfRAG 2026.04		69.1
GuarantRAG 2026.04		68.9
P-RAG 2026.04		68.1
DA-RAG 2026.04		67.3
SOLAR 2026.04		66
DRAGIN 2026.04		65.6
FLARE 2026.04		65.1
P-RAG 2026.04		64.7
RQ-RAG 2026.04		64.3
Standard RAG 2026.04		63.4
SOLAR 2026.04		63.1
DRAGIN 2026.04		62.8
SelfRAG 2026.04		62.5
FLARE 2026.04		62.4
RQ-RAG 2026.04		61.5
DA-RAG 2026.04		60.8
SelfRAG 2026.04		59.2
Qwen3-14B-T 2026.04		58.7
DA-RAG 2026.04		57.9
Standard RAG 2026.04		57.2
Standard RAG 2026.04		54.3
Qwen3-14B 2026.04		51.3
Qwen3-8B 2026.04		48.7