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

Citation and Evidence Recall on NeoQA

92.8Rk

GPT-5.2

4.60827.50450.473.296Oct 23, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.10
92.896.193.8
2025.10
8988.191
2025.10
88.58691.1
2025.10
77.157.282.2
2025.10
75.984.881.4
2025.10
75.671.773.2
2025.10
75.481.479.7
2025.10
74.577.480.8
2025.10
73.749.472.2
2025.10
70.810070.8
2025.10
70.167.275.2
2025.10
6810068
2025.10
63.655.461.6
2025.10
61.842.563
2025.10
61.552.861.4
2025.10
5147.652.5
2025.10
47.940.548.6
2025.10
45.548.353
2025.10
27.63735.7
2025.10
819.18.6