← Dictionary / Sequential Revision

Sequential Revision

Appears in 1 paper

A strategy where you iteratively refine a solution, using feedback from one attempt to improve the next.

As used in Paper 23 — Scaling LLM Test-Time Compute Optimally Can be More Effective than Scaling Model Parameters →

A strategy where you iteratively refine a solution, using feedback from one attempt to improve the next. Each round improves the solution further until convergence or a stopping criterion is reached. Particularly effective when the model can learn from its mistakes.

Appears in papers

Paper 23 — Scaling LLM Test-Time Compute Optimally Can be More Effective than Scaling Model Parameters →