Hint Tuning reduces token usage 24-66% (31.5% avg) in reasoning models via 1K self-annotated samples aligned to an instruct model's capabilities while keeping benchmark accuracy.
Training language models to reason efficiently
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
baseline 1
citation-polarity summary
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1roles
baseline 1polarities
baseline 1representative citing papers
citing papers explorer
-
Hint Tuning: Less Data Makes Better Reasoners
Hint Tuning reduces token usage 24-66% (31.5% avg) in reasoning models via 1K self-annotated samples aligned to an instruct model's capabilities while keeping benchmark accuracy.