Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-hop Question Answering on HotpotQA (F1)
Loading...
76.4
F1 Score
AdaptOrch
67.976
70.163
72.35
74.537
Feb 18, 2026
F1 Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
F1 Score
AdaptOrch
Lat.=1.3×, Tok(K)=22.7
2026.02
76.4
Self-MoA (matched)
Lat.=1.2×, Tok(K)=23.1
2026.02
75.5
Static-Parallel
Lat.=1.2×, Tok(K)=28.4
2026.02
72.8
MoA-3L
Lat.=2.5×, Tok(K)=47.3
2026.02
71.6
LLM-Blender
Lat.=1.5×, Tok(K)=34.8
2026.02
70.4
Static-Sequential
Lat.=2.1×, Tok(K)=26.1
2026.02
69.1
Single Best
Lat.=1.0×, Tok(K)=6.8
2026.02
68.3
Feedback
Search any
task
Search any
task