Rejection sampling: accept a draft token with probability min(1, p(x)/q(x))
cneuralnetwork · x · 2026-10-22
Part five covers verification via rejection sampling:
- A token x sampled from draft q is accepted with probability a(x) = min(1, p(x)/q(x));
- If p(x) ≥ q(x), acceptance is certain; otherwise it is probabilistic;
- Crucial detail: after rejection, you must not simply resample from p — that would break the distribution guarantee (the fix comes next).
Related event: Rejection and Residual Sampling Preserve Output Distribution(2 posts)→
More from Research
- Ask-to-Distill: synthesizing questions for answer-free on-policy distillation — UIUC-CS · 2026-10-22
- RFD4 team swaps self-conditioning for recurrent updates, echoing looped flows — AllThingsApx · 2026-10-22
- MIT ML technique breaks records for 2-6 week weather forecasts by learning to correct model errors — LesterMackey · 2026-10-22
- Harvard undergrad's ML subseasonal weather forecasting work, started at Microsoft Research, lands in Nature — LesterMackey · 2026-10-22
- Chatlet: generative UI that emerges from group chats to power collaborative decisions — HaijunXia · 2026-10-22
- If your research idea is concrete enough, agents can mostly execute it, says AI practitioner — burny_tech · 2026-10-22