List your company

Scientific Reasoning on Scientific Reasoning Tasks Agentic Subset

56.79Avg@8Agentic Self-Instruct
52.21453.40254.5955.778Jun 24, 2026

Evaluation results

MethodLinks
Agentic Self-InstructTraining Data Source=Agentic Self-Instruct, Base Model=Qwen3.5-4B2026.0656.7981.864.4
CoT Self-InstructTraining Data Source=CoT Self-Instruct, Base Model=Qwen3.5-4B2026.0656.3381.743.94
Combined (2× data)Training Data Source=Combined, Training Data Size=2× data, Base Model=Qwen3.5-4B2026.0655.8882.113.49
Qwen3.5-4BTraining Data Source=Base, Base Model=Qwen3.5-4B2026.0652.3981.13—