List your company

RLHF on HH-RLHF (test)

88.94Average ScoreGEOALIGN
41.214453.604765.99578.3853Jun 25, 2026

Evaluation results

MethodLinks
GEOALIGNBackbone=Qwen3-4B2026.0688.94
GEOALIGNBackbone=Qwen3-1.7B2026.0688.85
PF-PPO (BR)Backbone=Qwen3-4B2026.0687.71
PARBackbone=Qwen3-4B2026.0687.61
SEED-GRPOBackbone=Qwen3-4B2026.0687.43
BASE-GRPOBackbone=Qwen3-4B2026.0686.72
PF-PPO (BW)Backbone=Qwen3-4B2026.0686.6
PODSBackbone=Qwen3-4B2026.0685.94
PF-PPO (BR)Backbone=Qwen3-1.7B2026.0685.56
PARBackbone=Qwen3-1.7B2026.0684.91
SEED-GRPOBackbone=Qwen3-1.7B2026.0684.24
PF-PPO (BW)Backbone=Qwen3-1.7B2026.0683.96
BASE-GRPOBackbone=Qwen3-1.7B2026.0683.54
PODSBackbone=Qwen3-1.7B2026.0683.39
SFTBackbone=Qwen3-4B2026.0644.73
SFTBackbone=Qwen3-1.7B2026.0643.05