DLLM-JEPA: Joint Embedding Predictive Architectures for Masked Diffusion Language Models

ArXi:2606.00091v1 Announce Type: cross Joint Embedding Predictive Architectures (JEPAs) have reshaped self-supervised representation learning in vision. The recent LLM-JEPA ported JEPA to autoregressive language models but inherited two steep costs from the causal-attention substrate: it demands explicit multi-view data (e.g., text-code pairs), and it requires two gradient-carrying forward passes per step. We