Contrastive Decoding for Reasoning
Free while signed in. Answers cite the passages they came from.
Shows that contrastive decoding, a simple inference-time technique, substantially improves reasoning in large LLMs.
Contrastive decoding: Subtracts the log-probabilities of a smaller "expert" model from those of the target LLM, boosting tokens where the larger model confidently differs from the smaller one.
Llama 65B beats Llama 2: Contrastive decoding lets Llama 65B outperform Llama 2 and other strong baselines on commonsense and reasoning benchmarks.
Training-free: Requires no additional training - just a smaller model available at inference time and a modified decoding rule.
Generalizable lever: Positions contrastive decoding as a simple, cheap lever for reasoning improvement that can complement other prompting or fine-tuning techniques.
Get next week’s papers.
The same picks and the same summaries, in your inbox. Free, and 176 issues deep.
Subscribe on Substack