Hint Tuning: Less Data Makes Better Reasoners

ArXi:2605.08665v2 Announce Type: replace Large reasoning models achieve high accuracy through extended chain-of-thought but generate 5--8 tokens than necessary, applying verbose reasoning uniformly regardless of problem difficulty. We propose Hint Tuning, a data-efficient approach that teaches models to calibrate reasoning depth. Our key insight: the corresponding instruct model serves as an ideal difficulty probe. By testing what the instruct model can solve with varying guidance, we automatically construct.