CaLR: Causal Latent Revision for Robust Diffusion Reasoning

事件

arXiv:2609.20981v1 Announce Type: new Abstract: Autoregressive (AR) models suffer from local greediness, while diffusion language models (DLMs) often lack the strict causal structure required for reasoning. To combine the advantages and overcome the drawbacks of the dual, we propose Causal Latent Re

来源

本条目由采集管线自动抓取并发布，完整内容见下方来源链接。

*采集源：arXiv cs.AI*