List your company

Science Question Answering on Physics (mean@16, maj@16)

81.6Mean@16SC-SDPO
56.74463.19769.6576.103May 26, 2026

Evaluation results

MethodLinks
SC-SDPOBackbone=Qwen3-8B, Alpha=0.52026.0581.684.9
SDPOBackbone=Qwen3-8B, Filtering Strategy=Hard Filter2026.0576.479.6
SDPOBackbone=Qwen3-8B2026.0575.878.5
SC-SDPOBackbone=Qwen3-8B, Alpha=12026.0575.477.3
SRPOBackbone=Qwen3-8B2026.0573.476.3
SDPOBackbone=Qwen3-8B, Weighting Strategy=PACED2026.0569.271.7
GRPOBackbone=Qwen3-8B, Normalization=None2026.0569.176.5
GRPOBackbone=Qwen3-8B2026.0564.768.3
Qwen3-8BBackbone=Qwen3-8B2026.0557.753.8