ResearchResearch paperLarge Language Models · Image, Video & 3D Generation · Efficiency & Inference2 sources · Oct 13, 2026

Diff Diffusion: Predicting Edit Paths with Diffusion Models for Efficient Code Editing

We propose Diff Diffusion, a diffusion-inspired LLM framework for code editing that predicts edit paths from an original snippet to its revised form.

Diffusion large language models (dLLMs) are appealing for code generation because they can update many tokens in parallel, yet many are optimized for sequence synthesis rather than edit prediction.
Diff Diffusion represents denoising as structured edit operations, and combines this formulation with multi-stage training and efficient parallel decoding.
By focusing computation on changed regions, Diff Diffusion supports accurate low-latency edits and multi-position parallel modification.
Across public edit-completion benchmarks, edit-prediction evaluations, and online validation, Diff Diffusion matches or leads autoregressive baselines while delivering 2.78 times faster inference on public edit-completion benchmarks and a 39.0% P50 model-latency reduction in production.

[1]Diff Diffusion: Predicting Edit Paths with Diffusion Models for Efficient Code Editing
arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 13, 06:49 AM
We propose Diff Diffusion, a diffusion-inspired LLM framework for code editing that predicts edit paths from an original snippet to its revised form.
Diffusion large language models (dLLMs) are appealing for code generation because they can update many tokens in parallel, yet many are optimized for sequence synthesis rather than edit prediction.
[2]Diff Diffusion: Predicting Edit Paths with Diffusion Models for Efficient Code Editing
Hugging Face Daily Papers · Oct 13, 12:00 AM · same content

Extractive summary: sentences quoted from the sources.

Before this