Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-hop Retrieval on Morehopqa (test)
Loading...
80.5
Recall
ThinkGR
42.2696
52.1948
62.12
72.0452
May 21, 2026
Recall
Updated 12d ago
Evaluation Results
Method
Method
Links
Recall
ThinkGR
Model Parameters=8B
2026.05
80.5
w/o Thought
Model Parameters=8B
2026.05
78
GritHopper
Model Parameters=7B
2026.05
74.82
w/o RL
Model Parameters=8B
2026.05
73.84
R3-RAG
Model Parameters=8B
2026.05
70.44
RT-RAG
Model Parameters=8B
2026.05
66.95
IRCoT
Model Parameters=70B
2026.05
66.82
ITER-RETGEN
Model Parameters=70B
2026.05
60.73
Auto-RAG
Model Parameters=7B
2026.05
59.48
Selfask
Model Parameters=70B
2026.05
57.6
w/o SFT
Model Parameters=8B
2026.05
55.14
MDR
Model Parameters=110M
2026.05
49.6
BGE-large
Model Parameters=326M
2026.05
47.58
SEAL
Model Parameters=406M
2026.05
47.27
Contriever
2026.05
45.04
BM25
2026.05
43.74
Feedback
Search any
task
Search any
task