One Proposal for Every Margin: Zero-Shot Amortized Sequential Importance Sampling for Binary Matrices
MarginFlow learns a zero-shot sequential importance sampling proposal, matching or beating 31 analytic designs on 1187 of 1190 margins.
Stories per day · 30 days
3 stories, newest first · one card per story
Stories tagged #importance-sampling
Rethinking Training-Inference Mismatch in LLM Reinforcement Learning: Where It Arises and How to Correct It
Calibrated importance sampling corrects token-probability mismatch between LLM training and inference engines.
Robust Policy Optimization via Adversarial Importance Sampling
Adversarial Importance Sampling estimates worst-case RL returns without extra interactions; authors also release the advrl PyTorch library.