List your company

General Performance Evaluation on Aggregate (IFEval, GPQA, LCB, Arena-Hard, CW, MT-Bench, WildBench)

63.69Average ScoreSPARD
39.572445.833752.09558.3563Apr 9, 2026

Evaluation results

MethodLinks
SPARDBackbone=Qwen3-8B2026.0463.69
+ GRPOavgBackbone=Qwen3-8B2026.0462.17
+ GRPOimpBackbone=Qwen3-8B2026.0461.74
+ DPOBackbone=Qwen3-8B2026.0461.2
+ GRPOrmBackbone=Qwen3-8B2026.0461.15
BaseBackbone=Qwen3-8B2026.0460.56
SPARDBackbone=Qwen2.5-7B-Instruct2026.0450.03
+ SFTBackbone=Qwen3-8B2026.0449.6
+ GRPOavgBackbone=Qwen2.5-7B-Instruct2026.0448.46
+ DPOBackbone=Qwen2.5-7B-Instruct2026.0447.55
+ GRPOrmBackbone=Qwen2.5-7B-Instruct2026.0447.06
BaseBackbone=Qwen2.5-7B-Instruct2026.0446.12
+ GRPOimpBackbone=Qwen2.5-7B-Instruct2026.0445.59
+ SFTBackbone=Qwen2.5-7B-Instruct2026.0440.5