AI RESEARCH
Do Not Waste Your Rollouts: Recycling Search Experience for Efficient Test-Time Scaling
arXiv CS.LG
•
ArXi:2601.21684v2 Announce Type: replace-cross Test-Time Scaling enhances the reasoning capabilities of Large Language Models by allocating additional inference compute to broaden the exploration of the solution space. However, existing search strategies typically treat rollouts as disposable samples, where valuable intermediate insights are effectively discarded after each trial. This wasted rollout-level experience leads to substantial computational redundancy, as models repeatedly re-derive discovered