Untangling Fine-Grained Code Changes
Not on the map. Why?
Abstract (arXiv)
After working for some time, developers commit their code changes to a version control system. When doing so, they often bundle unrelated changes (e.g., bug fix and refactoring) in a single commit, thus creating a so-called tangled commit. Sharing tangled commits is problematic because it makes review, reversion, and integration of these commits harder and historical analyses of the project less reliable. Researchers have worked at untangling existing commits, i.e., finding which part of a commit relates to which task. In this paper, we contribute to this line of work in two ways: (1) A publicly available dataset of untangled code changes, created with the help of two developers who accurately split their code changes into self contained tasks over a period of four months; (2) a novel approach, EpiceaUntangler, to help developers share untangled commits (aka. atomic commits) by using fine-grained code change information. EpiceaUntangler is based and tested on the publicly available dataset, and further evaluated by deploying it to 7 developers, who used it for 2 weeks. We recorded a median success rate of 91% and average one of 75%, in automatically creating clusters of untangled fine-grained code changes.
Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). First posted on arXiv 24 Feb 2015.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2022: 1; 2023: 2; 2024: 0; 2025: 2; 2026: 5.
10 papers.
- 2022RACE: Retrieval-Augmented Commit Message Generationcited by 27
- 2026Are Coding Agents Generating Over-Mocked Tests? An Empirical Studycited by 5
- 2023CCRep: Learning Code Change Representations via Pre-Trained Code Model and Query Backcited by 3
- 2025LLM-Driven Collaborative Model for Untangling Commits via Explicit and Implicit Dependency Reasoningcited by 2
- 2025CodeFuse-CommitEval: Towards Benchmarking LLM's Power on Commit Message and Code Change Inconsistency Detectioncited by 1
- 2026Rethinking Code Review in the Age of AI: A Vision for Agentic Code Reviewcited by 1
- 2026"Refactoring Runaway": Understanding and Mitigating Tangled Refactorings in Coding Agents for Issue Resolutioncited by 1
- 2026SemaDiff: Identifying Semantic-Changing Commits with Generated Code and Testscited by 1
- 2023ActiveClean: Generating Line-Level Vulnerability Data via Active Learningcited by 0
- 2026DepRadar: Agentic Coordination for Context Aware Defect Impact Analysis in Deep Learning Librariescited by 0
- 2026SemaDiff: Identifying Semantic-Changing Commits with Generated Code and Tests14 Jul 2026
- 2026"Refactoring Runaway": Understanding and Mitigating Tangled Refactorings in Coding Agents for Issue Resolution21 May 2026
- 2026Rethinking Code Review in the Age of AI: A Vision for Agentic Code Review17 May 2026
- 2026Are Coding Agents Generating Over-Mocked Tests? An Empirical Study30 Jan 2026
- 2026DepRadar: Agentic Coordination for Context Aware Defect Impact Analysis in Deep Learning Libraries14 Jan 2026
- 2025CodeFuse-CommitEval: Towards Benchmarking LLM's Power on Commit Message and Code Change Inconsistency Detection25 Nov 2025
- 2025LLM-Driven Collaborative Model for Untangling Commits via Explicit and Implicit Dependency Reasoning22 Jul 2025
- 2023ActiveClean: Generating Line-Level Vulnerability Data via Active Learning4 Dec 2023
- 2023CCRep: Learning Code Change Representations via Pre-Trained Code Model and Query Back8 Feb 2023
- 2022RACE: Retrieval-Augmented Commit Message Generation5 Mar 2022