CoChat

Towards a Feasible Evaluation Function for Search-Based Merge Conflict Resolution

de Souza Campos, Heleno, de Menezes, Gleiph Ghiotto L., de Oliveira Barros, Márcio, van der Hoek, André, Gresta Paulino Murta, Leonardo · 2026 · ACM Transactions on Software Engineering and Methodology

The DOI matches its Crossref record; 27 of the 27 references it cites were found in a registry or an index.

Abstract

Resolving merge conflicts manually is a tedious and complex task. While automated approaches exist, many challenges persist. One promising yet underexplored solution is the use of search-based optimization algorithms, which require an evaluation function to measure the quality of intermediary solutions. However, using code compilation and test execution for this purpose is computationally expensive. This study investigates the relationship between conflict resolutions and conflicting content to identify a metric for guiding search-based optimization techniques. We analyzed 9,998 conflict chunks from 1,062 open source projects, focusing on the similarity of resolutions to their parents and the correlation between randomly generated candidates, parent versions, and the resolution. Our findings reveal that conflict resolutions are, on average, 70% similar to both parents. A strong median correlation ( \(\rho=0.791\) ) exists between candidate-parent and candidate-resolution similarities when aggregating parent similarities with the mean function. Based on these findings, we propose and evaluate SBCR, a Search-Based Conflict Resolution approach that uses parent similarity as a guiding function. We found that the resolution candidates generated by SBCR have a median of 86.5% similarity to the expected resolution, achieving 100% of similarity in 25.2% of the conflicts.

Overview

Integrity signals

Registry

Crossref · 10.1145/3748256

Cites

27 of 27 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Verification

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Citations

Cites

Show all 27 references

Research works printed but not linked: 2

  1. 10.5555/2486788.2486884 · doi:10.5555/2486788.2486884
  2. John W. Ratcliff and David E. Metzener. 1988. Pattern matching: The gestalt approach. Dr. Dobb’s Journal 13, 7 (1988), 46.

Cited by

Metadata

Authors
de Souza Campos, H., de Menezes, G. G. L., de Oliveira Barros, M., van der Hoek, A., & Gresta Paulino Murta, L.
Affiliations
  • Universidade Federal Fluminense, Niterói, Brazil
  • Universidade Federal de Juiz de Fora, Juiz de Fora, Brazil
  • Universidade Federal do Estado do Rio de Janeiro, Rio de Janeiro, Brazil
  • University of California, Irvine, Irvine, California, USA
Venue
ACM Transactions on Software Engineering and Methodology
Published
2026-04-24
Volume
35
Issue
5
Pages
1-35
ISSN
  • 1049-331X (ISSN-L)
  • 1557-7392
OpenAlex
W4412400182
Cited by
1
References
27
Citations by year
2026: 1
FWCI
0.64
Type
journal article
Language
en
Open access
closed
Locations
ACM Transactions on Software Engineering and Methodology (publishedVersion)
Funders
  • Conselho Nacional de Desenvolvimento Científico e Tecnológico (309300/2023-1)
  • National Science Foundation (CCF-2210812)
  • Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro (E-26/210.542/2024, E-26/204.145/2024)
Topics
  • Physical Sciences › Computer Science › Information Systems › Software Engineering Research
  • Physical Sciences › Computer Science › Information Systems › Software Engineering Techniques and Practices
  • Physical Sciences › Computer Science › Software › Software Testing and Debugging Techniques
Keywords
merge conflict resolution, evaluation function, search-based optimization, version control, open source projects

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.