Explore Broadly, Reason Sharply: Push Small Models toward the Frontier via Sampling
Parallel Power Tempering improves inference-time reasoning, matching frontier models without RL post-training.
2 stories, newest first · one card per story
Parallel Power Tempering improves inference-time reasoning, matching frontier models without RL post-training.
Parallel Power Tempering improves inference-time power sampling so small models match frontier reasoning without RL updates.
Today 1.6K readers · 2.5K views · 779 reads · all time 189K views