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

Social fairness evaluation on FACET (test)

92.58Accuracy

MS-PoE + BAIR

Updated 26d ago

Evaluation Results

Method	Links
MS-PoE + BAIR 2026.05		92.58	70.71	3.53
MS-PoE + BAIR 2026.05		91.94	51.58	3.7
MS-PoE 2026.05		91.82	51.05	3.78
Standard RAG + BAIR 2026.05		91.02	30.95	2.48
MAD-RAG + BAIR 2026.05		90.96	31.68	2.63
Standard RAG 2026.05		90.89	31.05	2.64
MAD-RAG 2026.05		90.86	30.84	2.65
Visual-focus Instr. + BAIR 2026.05		90.81	30.84	2.71
Visual-focus Instr. 2026.05		90.47	28.53	2.83
Baseline 2026.05		90.24	-	-
Standard RAG + BAIR 2026.05		88.56	50.31	4.64
LongLLMLingua + BAIR 2026.05		87.8	33.79	6.36
LongLLMLingua 2026.05		87.76	34.21	6.45
LongLLMLingua + BAIR 2026.05		87.45	45.97	4.89
Visual-focus Instr. + BAIR 2026.05		87.28	44.93	5.2
MAD-RAG + BAIR 2026.05		85.1	36.08	6.19
Baseline 2026.05		84.91	-	-
MAD-RAG 2026.05		84.41	25.53	5.11
Visual-focus Instr. 2026.05		84.38	28.25	5.65
LongLLMLingua 2026.05		84.36	31.59	6.26
Standard RAG 2026.05		84.34	26.62	5.41
MS-PoE 2026.05		82	29.07	8.6