Decoding in Order-Agnostic Language Models: Chain-Rule Deviation and Uniform Spreading 事件

Decoding in Order-Agnostic Language Models: Chain-Rule Deviation and Uniform Spreading arXiv:2606.00997v1 Announce Type: new Abstract: Order-agnostic language models (OALMs), including discrete diffusion language models (dLLMs), are trained to predict masked tokens under arbitrary conditioning sets, allowing sequences to be generated or scored under arbitrary reveal orders at inference time. In LLaDA-2.1, we report three findings. First, the learned conditionals are not exact factorizations of