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

Personalized Response Generation on AIME Implicit Preference 2025 (test)

0.221Preference Score

TriAlign

0.152360.170180.1880.20582Jun 1, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.06
0.2210.5110.4670.0250.0440.049
2026.06
0.2190.4080.3330.0590.0750.145
2026.06
0.2170.3940.2670.0730.1280.185
2026.06
0.1990.4580.3670.050.0910.109
2026.06
0.1840.4960.40.0770.0960.156
2026.06
0.1810.4320.3670.050.0660.117
0.170.4770.4090.0520.0680.109
2026.06
0.1550.4940.4230.0740.0710.149