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

Multi-hop Reasoning on MuSiQue IRCoT 500 samples (test)

25.59ACC

ActiShade

2.92848.811714.69520.5783Jan 12, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.01
25.5927.47
2026.01
25.2526.94
2026.01
22.826.11
2026.01
22.724.11
2026.01
21.1122.61
2026.01
20.621.41
2026.01
19.822.01
2026.01
19.7420.5
2026.01
18.4821.75
2026.01
18.2420.59
2026.01
17.3520.6
2026.01
17.221.54
2026.01
15.8919.9
2026.01
15.5718.32
2026.01
15.418.07
2026.01
14.918.01
2026.01
14.819.16
2026.01
13.118
2026.01
11.6516.29
2026.01
11.617.24
2026.01
11.4216.06
2026.01
10.6914.89
2026.01
10.417.74
2026.01
6.213.48
2026.01
613.93
2026.01
5.69.96
2026.01
3.811.09