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

Personalized Response Generation on StereoSet Implicit Preference (test)

0.474Pref Score

TriAlign

Updated 1d ago

Evaluation Results

Method	Links
TriAlign 2026.06		0.474	0.792	0.778	0.007	0.014	0.029
P-DEBIAS 2026.06		0.43	0.414	0.388	0.019	0.026	0.046
Qwen3-4B-Instruct-2507 2026.06		0.429	0.46	0.421	0.022	0.039	0.048
P-DEFENSE 2026.06		0.422	0.446	0.416	0.014	0.03	0.032
2STEPPROMPT 2026.06		0.422	0.355	0.342	0.007	0.014	0.019
SWIFT 2026.06		0.422	0.718	0.697	0.016	0.02	0.023
BESTPERSONA 2026.06		0.409	0.468	0.454	0.011	0.014	0.024
SFT 2026.06		0.409	0.668	0.653	0.008	0.014	0.012