Submission Link: https://icse2026-llm4code.hotcrp.com/

1. Paper Types and Formats

This workshop accepts research papers, position papers, and extended abstracts:

  • Research papers (4 – 8 pages including references) on novel approaches, tools, datasets, or studies.
  • Position papers (1 – 4 pages including references) on novel ideas and positions that have yet to be fully developed.
  • Extended abstracts (1 – 5 pages including references) that are free of article processing charges.

Important Note on Publication Charges: All ACM papers, excluding “extended abstracts” (up to five pages), are subject to ACM Open Access policies. ACM papers, excluding “extended abstracts”, require either an Open Access institutional agreement or article processing charges (APCs) paid by the authors (USD 250 for ACM members, USD 350 for non-members) for publication.

All papers must be in English, in PDF format, and must not exceed the page limits listed above. Papers should be submitted via HotCRP and will be reviewed in a double-blind process.

Submissions should use the official “ACM Primary Article Template”, which can be obtained from the ACM Proceedings Template page. LaTeX users should use the sigconf option and the review option (to produce line numbers for easy reference by reviewers). The following LaTeX code should be placed at the start of the document:

\documentclass[sigconf,review]{acmart}

Detailed submission policies and formatting guidelines are aligned with the ICSE 2026 Research Track submission process.

2. Archival vs. Non-Archival

All accepted papers by default will appear in the ICSE 2026 workshop proceedings (i.e., the archival option). The official publication date of the workshop proceedings is the date the proceedings are made available by ACM. This date may be up to two weeks prior to the first day of ICSE 2026. The official publication date affects the deadline for any patent filings related to published work.

We also provide a non-archival option on the paper submission system (HotCRP) for authors who prefer not to have their papers in the proceedings. For non-archival papers, the camera-ready version will only be posted on our workshop website (and will not be listed on DBLP).

Please note that regardless of which option you choose (archival or non-archival), submissions must be fully original work (not accepted or published elsewhere) at the time of submission, and at least one author of each accepted paper must register for the workshop and present the paper.

3. Best Paper Awards

🏆 We will announce best paper awards (up to 10% of accepted papers) during the workshop!

4. Topics

This workshop accepts submissions on the following topics (but is not limited to):

  • LLM Agents for code-relevant tasks
  • LLM applications on code-relevant tasks
    • LLM-based code generation
    • LLM-based fuzzing and testing
    • LLM-based test generation
    • LLM-based GUI testing
    • LLM-based mobile application testing
    • LLM-based fault localization
    • LLM-based program repair
    • LLM-based vulnerability detection
    • LLM-based code maintenance
    • LLM-based program analysis
    • LLM-based code comprehension
    • LLM-based reverse engineering
    • LLM-based code evolution and maintenance
    • LLM-based refactoring
  • Datasets and Evaluation
    • Datasets for LLM4Code pre-training
    • Datasets for LLM4Code post-training
    • Datasets for LLM4Code evaluation
    • Automated dataset generation/augmentation (e.g., via LLMs)
    • Empirical studies on LLM4Code
  • Model design and optimization on LLM4Code
    • Model architecture design
    • Model hyperparameter tuning
    • Prompt tuning and prompt engineering
    • Pretraining objective design
    • Model alignment
    • Model distillation
    • Model optimization/quantization