Revolutionizing Decoding in Diffusion Language Models
New algorithms transform constrained decoding for diffusion language models, boosting accuracy with minimal overhead. Is this the future of inference?
Read the full articleYou might also wanna read
Fast-dLLM: Training-Free Acceleration Method for Diffusion Language Models Using KV Cache and Parallel Decoding
Diffusion-based large language models (Diffusion LLMs) have shown promise for non-autoregressive text generation with parallel decoding capa
arxiv.org·8mo ago
[ICML 2026] The Flexibility Trap: Rethinking the Value of Arbitrary Order in Diffusion Language Models
Authors: Zanlin Ni, Shenzhi Wang, Yang Yue, Tianyu Yu, Weilin Zhao, Yeguo Hua, Tianyi Chen, Jun Song, Cheng Yu, Bo Zheng, Gao Huang
Three training-time interventions improve diffusion-based speculative decoding by 21-76%
Large language models (LLMs) achieve remarkable performance across a wide range of tasks, but their autoregressive decoding process incurs s
arxiv.org·15d ago
Google's DiffusionGemma achieves 4x faster text generation using diffusion-based approach
An overview of DiffusionGemma, an exceptionally fast text generation model with up to 4x faster speeds.
deepmind.google·25d ago
Google's DiffusionGemma achieves 4x faster text generation using diffusion-based parallel token generation
An overview of DiffusionGemma, an exceptionally fast text generation model with up to 4x faster speeds.
blog.google·1mo agoGoogle's DiffusionGemma achieves 4x faster text generation using diffusion-based parallel token generation
An overview of DiffusionGemma, an exceptionally fast text generation model with up to 4x faster speeds.
blog.google·1mo ago
Consistency Diffusion Language Models Achieve 14x Faster Inference Through KV Caching and Step Reduction
Standard diffusion language models can't use KV caching and need too many refinement steps to be practical. CDLM fixes both with a post-trai
Together·4mo agoConsistency Diffusion Language Models Achieve 14x Faster Inference Through KV Caching and Step Reduction
Standard diffusion language models can't use KV caching and need too many refinement steps to be practical. CDLM fixes both with a post-trai
together.ai·4mo ago
Comments
Sign in to join the conversation.
No comments yet. Be the first.