NES: An Instruction-Free, Low-Latency Next Edit Suggestion Framework Powered by Learned Historical Editing Trajectories

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chen, Xinfang, Xiao, Siyang, Zhu, Xianying, Xie, Junhong, Liang, Ming, Chen, Dajun, Jiang, Wei, Li, Yong, Di, Peng
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866918422220111872
author Chen, Xinfang
Xiao, Siyang
Zhu, Xianying
Xie, Junhong
Liang, Ming
Chen, Dajun
Jiang, Wei
Li, Yong
Di, Peng
author_facet Chen, Xinfang
Xiao, Siyang
Zhu, Xianying
Xie, Junhong
Liang, Ming
Chen, Dajun
Jiang, Wei
Li, Yong
Di, Peng
contents Code editing is a frequent yet cognitively demanding task in software development. Existing AI-powered tools often disrupt developer flow by requiring explicit natural language instructions and suffer from high latency, limiting real-world usability. We present NES (Next Edit Suggestion), an instruction-free, low-latency code editing framework that leverages learned historical editing trajectories to implicitly capture developers' goals and coding habits. NES features a dual-model architecture: one model predicts the next edit location and the other generates the precise code change, both without any user instruction. Trained on our open-sourced SFT and DAPO datasets, NES achieves state-of-the-art performance (75.6% location accuracy, 27.7% exact match rate) while delivering suggestions in under 250ms. Deployed at Ant Group, NES serves over 20,000 developers through a seamless Tab-key interaction, achieving effective acceptance rates of 51.55% for location predictions and 43.44% for edits, demonstrating its practical impact in real-world development workflows.
format Preprint
id arxiv_https___arxiv_org_abs_2508_02473
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle NES: An Instruction-Free, Low-Latency Next Edit Suggestion Framework Powered by Learned Historical Editing Trajectories
Chen, Xinfang
Xiao, Siyang
Zhu, Xianying
Xie, Junhong
Liang, Ming
Chen, Dajun
Jiang, Wei
Li, Yong
Di, Peng
Software Engineering
Machine Learning
68N30
D.2.3; D.1.2; I.2.2
Code editing is a frequent yet cognitively demanding task in software development. Existing AI-powered tools often disrupt developer flow by requiring explicit natural language instructions and suffer from high latency, limiting real-world usability. We present NES (Next Edit Suggestion), an instruction-free, low-latency code editing framework that leverages learned historical editing trajectories to implicitly capture developers' goals and coding habits. NES features a dual-model architecture: one model predicts the next edit location and the other generates the precise code change, both without any user instruction. Trained on our open-sourced SFT and DAPO datasets, NES achieves state-of-the-art performance (75.6% location accuracy, 27.7% exact match rate) while delivering suggestions in under 250ms. Deployed at Ant Group, NES serves over 20,000 developers through a seamless Tab-key interaction, achieving effective acceptance rates of 51.55% for location predictions and 43.44% for edits, demonstrating its practical impact in real-world development workflows.
title NES: An Instruction-Free, Low-Latency Next Edit Suggestion Framework Powered by Learned Historical Editing Trajectories
topic Software Engineering
Machine Learning
68N30
D.2.3; D.1.2; I.2.2
url https://arxiv.org/abs/2508.02473