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

Multi-hop Question Answering on MuSiQue (Rubric Judge Accuracy)

39.6Rubric Judge Accuracy

Predicate-Based Belief State

7.77616.03824.332.562May 7, 2026
Updated 24d ago

Evaluation Results

MethodLinks
2026.05
39.6
2026.05
37.1
2026.05
36.8
2026.05
34.4
2026.05
33.8
2026.05
31.4
2026.05
30.6
2026.05
30.1
2026.05
29.6
2026.05
29.6
2026.05
25.9
2026.05
21.1
2026.05
14.4
2026.05
12.2
2026.05
9