Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
End-to-end question-answering on HotPotQA (test)
Loading...
0.464
EM
INTRA
0.33712
0.37006
0.403
0.43594
May 7, 2026
EM
Updated 26d ago
Evaluation Results
Method
Method
Links
EM
INTRA
Retrieval method=INTRA...
2026.05
0.464
Hybrid RAG
Retrieval method=Hybri...
2026.05
0.434
BGE
Retrieval method=BGE,...
2026.05
0.419
Qwen3-Emb-4B + Reranker
Retrieval method=Qwen3...
2026.05
0.416
MaxSim
Retrieval method=MaxSi...
2026.05
0.407
BM25
Retrieval method=BM25,...
2026.05
0.405
Qwen3-Emb-4B
Retrieval method=Qwen3...
2026.05
0.403
Qwen3-Emb-0.6B
Retrieval method=Qwen3...
2026.05
0.37
TF-IDF
Retrieval method=TF-ID...
2026.05
0.342
Feedback
Search any
task
Search any
task