Second Workshop on Large Language Models For Generative Software Engineering (LLM4SE 2026)
The workshop is co-located with Software Technologies: Applications and Foundations (STAF2026)

website: https://conf.researchr.org/home/staf-2026/llm4se-2026

The introduction of ChatGPT, Google’s Gemini, and Bing Chat AI marked a significant advancement in the field of natural language processing, contributing to the widespread popularity of large language models (LLMs). These models have found successful applications in diverse sectors, including healthcare, finance, education, and various engineering fields, among others. Agents powered by LLMs, such as Claude Code and GitHub Copilot, have revolutionised the landscape of software engineering. They significantly enhance developers’ efficiency in several classical software engineering tasks like coding, testing, and documentation. That is, LLMs are transforming the way we interact with computers, and their application to software engineering holds immense potential.

This workshop will explore how these powerful models can be harnessed to automate, augment, and even transform the software development lifecycle, leading to increased productivity, innovative solutions, and a new era of software creation. In this scenario, the role of those who will use these powerful tools and how they will use them is of paramount importance. On the one hand, generative AI can assist in generating code; on the other hand, the conversion from natural language and the governance of these tools will be handled by software engineers through various techniques proposed by the AI community.

Hence, the primary objective of this full-day workshop is to explore potential scenarios where software engineers can adopt LLMs to perform various tasks and engage in discussions regarding the implementation of such solutions and the philosophical implications of this significant technological advancement

-----------------------
Topic of Interest
-----------------------

Topics of interest for the workshop include, but are not limited to:

- LLMs for Requirements Generation (e.g., requirements collection, organisation, analysis).
- LLMs for Verification and Validation (e.g., test case generation, model-based testing, functional testing).
- LLMs for Model-Driven Engineering.
- LLMs for Low-Code Generation.
- LLMs for Code Generation.
- LLMs for Code Summarization (e.g., description generation, comments generation).
- LLMs for Patch Generation (e.g., repair, bug detection).
- LLMs for Code Optimization (e.g., performance, energy).
- LLMs for Code Translation.
Quality of LLM-generated Software Artifacts.

Submission Info

Two kinds of papers are solicited:

- regular papers (10 pages + 1 page for references)
- short papers (4 pages + 1 page for references)

Contributions should present novel research ideas (even if at a preliminary development stage), challenging problems, and practical contributions to the domain. Industrial experience reports or case studies related to the development or use of LLMs in a broad software engineering context are also solicited. Furthermore, extended abstracts are accepted, that can include prior work or visionary work, but will not be included in the proceedings. All papers must be written in English adhere to the CEUR Style single-column formatting guidelines available here Template Link, and be submitted through EasyChair at the link: https://easychair.org/conferences/?conf=staf2026 and then selecting the track Large Language Models for Software Engineering (LLM4SE).

The guidelines for the CEUR submission format are available here https://ceur-ws.org/HOWTOSUBMIT.html. Please read the instructions carefully, especially concerning the copyright. For any doubt you can contact one of the organizers.

-----------------------
Important Dates
-----------------------

- Abstract Submission: March 20th, 2026 AoE
- Paper Submission: March 27th, 2026
- Author Notification: April 30th, 2026 AoE
- Camera-Ready: May 14th, 2026 AoE

------------------------------
Organizing Commitee
-----------------------------

Boqi Chen (McGill University, Canada)
Robert Clarisó (Universitat Oberta de Catalunya, Spain)
Antonio Cicchetti (Mälardalens University, Sweden)
Riccardo Rubei (Mälardalens University, Sweden)