Discrete Stochastic Localization for Non-autoregressive Generation
arXiv:2602.16169v1 Announce Type: new Abstract: Non-autoregressive (NAR) generation reduces decoding latency by predicting many tokens in parallel, but iterative refinement often suffers from error accumulation …
Yunshu Wu, Jiayi Cheng, Partha Thakuria, Rob Brekelmans, Evangelos E. Papalexakis, Greg Ver Steeg
5 views
