Feature Selection Evaluation on GPT-4 Feature Relevance Estimation Suite Silver Standard (test) benchmark leaderboard · SOTA2 Research

Feature Selection Evaluation on GPT-4 Feature Relevance Estimation Suite Silver Standard (test)

74T1D ScoreGAME

Evaluation Results

Method	Links
GAME2025.02		74	78.6	72.5	66.8	68.4	66.5	71.1
GAME (BGE)2025.02		74	76.7	71.7	65.3	68.7	67.6	70.6
Ablation (g)2025.02		74	76	67.8	64.3	67.4	65.2	69.1
GAME (CODER)2025.02		72.8	77	71.3	65.9	67.3	66.8	70.2
Ablation (e)2025.02		71.4	75.8	66.6	62.2	66.4	65.4	68
Ablation (f)2025.02		71.2	76.4	65.1	66.5	66.6	66.5	68.7
PPMI-SVD (MGB)2025.02		70.7	73.4	70.7	68.3	71.6	73.6	71.4
Ablation (a)2025.02		69.6	75.2	60	64.8	65.3	64.8	66.6
Ablation (d)2025.02		69.5	74.7	64.9	60.3	64.9	62.5	66.2
OpenAI2025.02		69.3	69.4	65.9	68.4	61.5	66.8	66.9
Ablation (b)2025.02		68.3	72.4	60.5	58.2	63.6	60.4	63.9
Ablation (c)2025.02		68.2	74	57	59.4	61.2	61.2	63.5
PPMI-SVD (VA)2025.02		68	71.4	74.4	59.7	65.8	67.4	67.8
PPMI-SVD (Bor)2025.02		66.8	73.6	65	62.7	63.9	67.1	66.5
Ablation (h)2025.02		65.8	71.6	61.7	61.7	62.2	57.9	63.5
CODER2025.02		64.4	59.8	63	61.9	53.4	55.4	59.7
PPMI-SVD (UPMC)2025.02		63.9	65.6	61.5	61.1	58.2	60.4	61.8
BGE2025.02		63	54.2	59.6	56.8	54.9	54.2	57.1
PPMI-SVD (MIMIC)2025.02		62.7	62.2	—	44.2	59.6	59	57.6
PPMI-SVD (AVE)2025.02		62.5	67.1	67.9	57.8	62.9	64.2	63
SBERT2025.02		56.8	46.6	54.6	53.3	52.5	50.7	52.4
PPMI-SVD (BCH)2025.02		56.3	66.5	—	62.1	62.6	62.7	62
PPMI-SVD (Duke)2025.02		49.1	56.9	—	46.4	58.4	59.2	54
PBERT2025.02		45.8	43.7	47.3	45.9	46.2	45.7	45.7
BBERT2025.02		44.7	44.2	46.9	45.8	43.1	49.1	45.6