arXiv:2606. 14620v1 Announce Type: new Abstract: Open diffusion language models are marketed as parallel, non-autoregressive decoders, yet the order in which a shipped checkpoint actually commits its tokens is almost never measured.
arXiv:2608. 05687v1 Announce Type: cross Abstract: Masked diffusion language models (dLLMs) can commit tokens in any order -- a freedom marketed as their core advantage over autoregressive decoding.
arXiv:2606. 11552v1 Announce Type: cross Abstract: Large language models (LLMs) achieve remarkable performance across a wide range of tasks, but their autoregressive decoding process incurs substantial inference costs due to inherently sequential token generation.
The paper studies how the order in which tokens are committed in masked diffusion language models affects accuracy. It finds that when the final answer is committed before the preceding reasoning (an answer‑first trajectory), accuracy can suffer compared to unrestricted decoding, especially on tasks like GSM8K and MATH‑500. Experiments with controlled token positions show that delaying the answer token can improve performance, indicating that commitment order influences the context and output allocation of the model.
arXiv:2607. 28166v2 Announce Type: replace-cross Abstract: Diffusion language models expose a provisional prediction at every denoising step, and on many tasks the candidate answer inside it stabilizes before the step schedule is exhausted.
arXiv:2609.38536v1 Announce Type: cross Abstract: Diffusion-based large language models (dLLMs) promise to break the sequential latency bottleneck of autoregressive agents through parallel decoding,...
arXiv:2606. 20560v1 Announce Type: cross Abstract: LLM reasoning transparency is a critical affordance for understanding model decisions, mitigating misuse and misalignment, and debugging surprising model behaviors.
Large language models (LLMs) achieve remarkable performance across a wide range of tasks, but their autoregressive decoding process incurs substantial inference costs due to inherently sequential token generation. Speculative decoding addresses this bottleneck by employing a lightweight draft model to propose multiple future tokens that are subsequently verified in parallel by a larger target model.
arXiv:2608.30427v1 Announce Type: cross Abstract: Speculative decoding speeds up generation with an efficient draft model (drafter) that proposes tokens for a target model to verify in one pass, pres...
The paper compares the parallelism capabilities of three diffusion large language model paradigms—masked, uniform, and Gaussian diffusion. It proves that uniform and Gaussian diffusion can sample with a number of forward passes scaling with the dual total correlation of the distribution, potentially much less than the context length, whereas masked diffusion may require more passes. The study establishes a provable separation in parallelism, showing that masked diffusion’s critical windows are asymptotically narrower than those of the other two approaches.
arXiv:2606. 02955v1 Announce Type: cross Abstract: Diffusion large language models promise parallel token generation, yet inference remains bottlenecked by deciding which masked tokens can be safely committed together.
arXiv:2608.20530v1 Announce Type: new Abstract: Speculative decoding accelerates language-model inference by drafting future tokens that the target model verifies in parallel. A diffusion-style block...
arXiv:2609.39560v1 Announce Type: new Abstract: Sampling several responses and voting over their answers can improve a language model's accuracy, but repeated answers limit the benefit of additional...