List your company

Science Question Answering on Chemistry

80.6Mean@16SC-SDPO
40.87251.18661.571.814May 26, 2026

Evaluation results

MethodLinks
SC-SDPOBackbone=Qwen3-8B, Alpha=0.52026.0580.681.3
SRPOBackbone=Qwen3-8B2026.0580.581.6
SDPOBackbone=Qwen3-8B, Filtering Strategy=Hard Filter2026.058080.4
SC-SDPOBackbone=Qwen3-8B, Alpha=12026.058079.5
SDPOBackbone=Qwen3-8B2026.0578.779.2
SDPOBackbone=Qwen3-8B, Weighting Strategy=PACED2026.0574.775.5
GRPOBackbone=Qwen3-8B2026.0568.570.9
GRPOBackbone=Qwen3-8B, Normalization=None2026.0563.967
Qwen3-8BBackbone=Qwen3-8B2026.0542.435.7