Adaptive Replay Refinement: Allocating Test-Time Compute Between Revision and Resampling

Key points

• Test-time scaling allocates additional inference compute to refining existing solutions or generating new ones.1
• This raises a fundamental question: Should I keep improving this solution, or stop working on it and try a completely new solution?1
• Across seven models from five families and seven benchmarks, ARR matches majority voting on three of five mathematical reasoning benchmarks at equal compute, while reducing critical-path serial passes by up to 2.4 times.1
• These results demonstrate that detecting refinement exhaustion enables adaptive reallocation from revision to independent sampling, improving inference efficiency while highlighting the limits of self-revision.1

Sources (1)

• [1]Adaptive Replay Refinement: Allocating Test-Time Compute Between Revision and Resampling
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 20, 02:41 PM

  We introduce Adaptive Replay Refinement (ARR), a training-free method that detects when a stationary revision reaches a fixed point and reallocates the released compute to a new independent sample; replay decoding treats the current solution as a draft and verifies unchanged tokens in parallel, enabling efficient fixed-point detection.

  Test-time scaling allocates additional inference compute to refining existing solutions or generating new ones.

Extractive summary: sentences quoted from the sources.