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

Multimodal Preference Evaluation on Polaris

57.8tau_c

Polos

52.0853.56555.0556.535Oct 1, 2025
Updated 17d ago

Evaluation Results

MethodLinks
2025.10
57.8-
2025.10
56-
2025.10
5477.5
2025.10
5479.5
2025.10
53.979.4
2025.10
52.577
2025.10
52.379.7
2025.10
52.360
2025.10
52.3-
2025.10
52.373.3