Self-Play Search Distillation for Large Language Model Reasoning
Self-play search distillation raises Qwen3-4B-Base's math benchmark mean from 24.1 to 36.6.
1 stories, newest first · one card per story
Self-play search distillation raises Qwen3-4B-Base's math benchmark mean from 24.1 to 36.6.
Today 223 readers · 239 views · 63 reads · all time 199K views