List your company

Question Answering on HeadQA (pass@1)

78.12Pass@1Llama3.1-8B-Instruct
73.33674.57875.8277.062Aug 25, 2025

Evaluation results

MethodLinks
Llama3.1-8B-InstructTraining Pipeline=PSFT → GRPO2025.0878.12
Llama3.1-8B-InstructTraining Pipeline=PSFT2025.0877.9
Qwen2.5-7B-InstructTraining Pipeline=PSFT → GRPO2025.0875.89
Qwen2.5-7B-InstructTraining Pipeline=PSFT2025.0875.82
Qwen2.5-7B-InstructTraining Pipeline=SFT → GRPO2025.0875.42
Qwen2.5-7B-InstructTraining Pipeline=SFT2025.0874.65
Llama3.1-8B-InstructTraining Pipeline=SFT → GRPO2025.0873.81
Llama3.1-8B-InstructTraining Pipeline=SFT2025.0873.52