Show HN: Training-free improvement of Pareto frontiers for reasoning LLMs

1 points | by CreativeOn 13 hours ago