List your company

Short-form Question Answering on 2wiki (test)

26.5EMPPO + PRS_short
8.206412.955717.70522.4543Dec 8, 2025

Evaluation results

MethodLinks
PPO + PRS_shortBackbone=Qwen2.5-3B-Instruct2025.1226.5
VSPO + PRS_shortBackbone=Qwen2.5-3B-Instruct2025.1226.5
GRPO + PRS_shortBackbone=Qwen2.5-3B-Instruct2025.1226
SFTBackbone=Qwen2.5-3B-Instruct2025.1224.8
UntrainedBackbone=Qwen2.5-3B-Instruct2025.128.91