DMax: Aggressive Parallel Decoding for dLLMs

[ICML 2026] [OnlineSPEC] When Drafts Evolve: Speculative Decoding Meets Online Learning

Repository of the paper "Accelerating Transformer Inference for Translation via Parallel Decoding"

[ICML 2026] Jacobi Forcing: Fast and Accurate Diffusion-style Decoding

[ICLR 2026 Oral] Locality-aware Parallel Decoding for Efficient Autoregressive Image Generation

[ICLR 2026] Autoregressive Image Generation with Randomized Parallel Decoding

[ICLR 2026] dParallel: Learnable Parallel Decoding for dLLMs

[ICLR 2026] ParallelBench: Understanding the Tradeoffs of Parallel Decoding in Diffusion LLMs

Implementation of Flash-DLM (paper: FlashDLM: Accelerating Diffusion Language Models via Efficient KV Caching and Guided Diffusion). Provides training-free methods to accelerate diffusion language model inference.

📚 A curated list of Awesome Efficient dLLMs Papers with Codes

This repository contains the source code for COrAL, an Order-Agnostic Language Modeling framework for Efficient Iterative Refinement.

Parallel Decoding of Language Models: A Survey. A systematic survey of parallel decoding, covering speculative decoding, diffusion LLMs, etc...

Diffusion VLM papers indexed with one feature schema, plus decode-behaviour traces measured through a single instrumentation path

What parallel commitment costs a masked diffusion LM: a 4-page note, measured on a task with a decidable verifier. Produced by the Rust stack (candle-mi -> canvas).

Pure golang package for reading and writing xz-compressed files

A curated list for Efficient Inference of Diffusion-based Large Language Models