Seventh International Workshop on Deep Learning for Testing and Testing for Deep Learning (DeepTest 2026)

Machine Learning (ML) is widely adopted in modern software systems, including safety-critical domains such as autonomous cars, medical diagnosis, and aircraft collision avoidance systems. Thus, it is crucial to rigorously test such applications to ensure high dependability. However, standard notions of software quality and reliability become irrelevant when considering ML systems, due to their non-deterministic nature and the lack of a transparent understanding of the models’ semantics. ML is also expected to revolutionize software development. Indeed, ML is being applied for devising novel program analysis and software testing techniques related to malware detection, bug-finding, and type-checking

DeepTest 2026 aims to bring together academics and industry experts to discuss practical solutions and build momentum in this rapidly evolving field. The workshop will include invited talks and research presentations, providing a platform for participants to exchange ideas and insights.

This edition of DeepTest will be co-located with ICSE 2026, taking place from April 12 to April 18, 2026, in Rio de Janeiro, Brazil. The workshop will take place on Monday, April 13th.

Previous Editions

• DeepTest 2025 was co-located with ICSE 2025
• DeepTest 2024 was co-located with ICSE 2024
• DeepTest 2023 was co-located with ICSE 2023
• DeepTest 2021 was co-located with ICSE 2021
• DeepTest 2020 was co-located with ICSE 2020
• DeepTest 2019 was co-located with ICSE 2019

Testing Large Language Models Competition: Benchmarking the “Talking Manual”

This year, we are excited to feature a Tool Competition on testing Large Language Models. More details on the “Call for Competition Submissions” tab.

The workshop is partially supported by the EU project Sec4AI4Sec.

Industrial research prototype for the tool competition provided by BMW Group.

Supporters

Industrial research prototype provider

Accepted Papers

Call for Papers

DeepTest is an interdisciplinary workshop targeting research at the intersection of software engineering and deep learning. This workshop will explore issues related to:

• Deep Learning applied to Software Engineering (DL4SE)
• Software Engineering applied to Deep Learning (SE4DL)

Although the main focus is on Deep Learning, we also encourage submissions that are more broadly related to Machine Learning.

Topics of Interest

We welcome submissions introducing technology (i.e., frameworks, libraries, program analyses and tool evaluation) for testing DL-based applications, and DL-based solutions to solve open research problems (e.g., what is a bug in a DL/RL model). Relevant topics include, but are not limited to:

• High-quality benchmarks for evaluating DL/RL approaches
• Surveys and case studies using DL/RL technology
• Techniques to aid interpretable DL/RL techniques
• Techniques to improve the design of reliable DL/RL models
• DL/RL-aided software development approaches
• DL/RL for fault prediction, localization and repair
• Fuzzing DL/RL systems
• Metamorphic testing as software quality assurance
• Fault Localization and Anomaly Detection
• Use of DL for analyzing natural language-like artefacts such as code, or user reviews
• DL/RL techniques to support automated software testing
• DL/RL to aid program comprehension, program transformation, and program generation
• Safety and security of DL/RL based systems
• New approaches to estimate and measure uncertainty in DL/RL models

Types of Submissions

We accept two types of submissions:

• Full research papers: up to 8-page papers (including references) describing original and unpublished results related to the workshop topics;
• Short papers: up to 4-page papers (including references) describing preliminary work, new insights in previous work, or demonstrations of testing-related tools and prototypes.
• Extended abstracts: up to 5 pages (including references) for position or experience pieces, including emerging results, lessons learned, open problems, or novel ideas. These submissions are exempt from Article Processing Charges (APCs) under the current ACM open-access policy.

Formatting and Submission

Please submit your paper as a PDF via our HotCRP portal: https://deeptest2026.hotcrp.com/.

All submissions must strictly follow the official ACM Primary Article Template. You can find the template on the ACM Proceedings Template Page. If you are a LaTeX user, please use the sigconf option. For the double-blind review process, also use the review option to include line numbers and the anonymous option to remove author names. Please be aware that any alterations to spacing, font size, or other formatting that deviate from these instructions may result in your submission being desk-rejected without further review.

Double-Blind Review: DeepTest 2026 uses a double-blind review process. To ensure this, your submission must not reveal the authors’ identities. Please make every effort to honor this policy by omitting author names from the paper.

Plagiarism: By submitting, you acknowledge and agree to be bound by the ACM Policy and Procedures on Plagiarism and the IEEE Plagiarism FAQ.

If you have any questions about the submission process or if your work is a good fit for the workshop, don’t hesitate to contact the organizers.

Call for Competition Submissions

Introduction

The competition focuses on evaluating LLM-based automotive assistants that help drivers interpret vehicle manuals. Because many vehicle functions, such as Adaptive Cruise Control, depend on specific safety-critical conditions, manuals include warnings that an assistant must correctly relay. Since LLMs can omit or hallucinate such warnings, the challenge aims to automatically stress-test assistants for their reliability in conveying accurate safety information.

Participants will compete on designing automated test generators that create diverse textual user requests intended to make the system fail, specifically, to trigger situations where required warnings are not included in the assistant’s response. Tools will be ranked based on metrics such as the number and diversity of warnings successfully ignored as well as the diversity of the generated test inputs, supporting better root-cause analysis.

All submissions will be evaluated on both a simplified public RAG-based information system and an industrial research prototype provided by BMW Group.

How to participate

All competition details, tasks, rules, and submission instructions can be found in our GitHub repository.

Important dates

• Registration deadline: 30/11/2025
• Submission deadline: 31/12/2025
• Results notification: 21/01/2026
• Presentation @ ICSE: 13/04/2026

We encourage everyone, students, academics, industry teams, and independent developers, to join and contribute to improving how we test and understand modern LLMs.