List your company

Scientific Reasoning on GPQA D (Avg@4)

34.7Avg@4GRPO
28.66830.23431.833.366May 8, 2026

Evaluation results

MethodLinks
GRPOBackbone=Llama3.2-3B-Instruct, Domain Type=In-domain2026.0534.7
Dr.GRPOBackbone=Llama3.2-3B-Instruct, Domain Type=In-domain2026.0534.3
PRIMEBackbone=Llama3.2-3B-Instruct, Domain Type=In-domain2026.0533.1
OriginalBackbone=Llama3.2-3B-Instruct, Domain Type=In-domain2026.0528.9