Researchers propose BlockServe, a continuous batching framework for diffusion large language models that uses block-grained scheduling to eliminate compute bubbles and tail latency caused by convergen…
Researchers propose Trajectory-Aware Commit Gating (TACG), a training-free decoder for diffusion language models that uses trajectory-aware signals to decide when to commit tokens, improving accuracy …
ByteDance and Renmin University researchers released iLLaDA, an 8B diffusion language model that matches Qwen2.5 on base benchmarks but lags after fine-tuning. The model, trained from scratch on 12 tr…
Researchers introduced iLLaDA, an 8B masked diffusion language model trained from scratch with fully bidirectional attention, scaling pre-training to 12T tokens and fine-tuning on a 25B-token instruct…
Researchers have developed Token-to-Mask (T2M) remasking, a training-free replacement for Token-to-Token (T2T) editing in discrete masked diffusion language models like LLaDA. The method resets suspec…