2026
Cache Coherent Resampling for Efficient Test Time Scaling in LLM Reasoning via Adaptive Sequential Monte Carlo
ICML 2026poster
Recent work shows that chain based sampling for power shaped trajectory distributions can deliver large test time gains from a fixed base LLM and can approach RL trained reasoners such as GRPO. Deployment is the bottleneck. Autoregressive Metropolis Hastings is inherently serial, limits GPU utilizat…