Share your thoughts, 1 month free Claude Pro on usSee more

Structured Reasoning (Code/SQL) on Structured OOD

92.1Full Accuracy

Qwen3-14B Base

Updated 5d ago

Evaluation Results

Method	Links
Qwen3-14B Base 2026.05		92.1	83.2	52.6
Qwen3-8B OPSD 2026.05		87.8	77.9	54.8
Qwen3-8B SFT 2026.05		86.9	76.9	64.4
Qwen3-8B GRPO 2026.05		86.9	84	45.8
Qwen3-8B CCOPD 2026.05		86.2	77.9	67.9
Qwen3-8B 4B-teacher CCOPD 2026.05		85.6	78.8	67.6
Qwen3-8B Base 2026.05		85.3	78.8	48.7
Qwen3-8B Forward-CCOPD 2026.05		84.6	76	65.4
Qwen3-8B 14B-teacher CCOPD 2026.05		84.6	78.2	66.3
Qwen3-4B CCOPD 2026.05		83.7	76	63.8
Qwen3-4B Base 2026.05		80.1	74	41
Llama3.1-8B Base 2026.05		67.6	63.7	37.5
Llama3.1-8B CCOPD 2026.05		67.3	64.1	52.7