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

Evidence Retrieval on G-bench Novel

87.7Recall

G-reasoner

65.23671.06876.982.732Sep 29, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.09
87.7
2025.09
82.6
2025.09
82.4
2025.09
82.1
2025.09
81.2
2025.09
79.6
2025.09
75.9
2025.09
67.4
2025.09
66.2
2025.09
66.1