MSR 2026
Mon 13 - Tue 14 April 2026 Rio de Janeiro, Brazil
co-located with ICSE 2026
Events (25 results)

Are We All Using Agents Now? An Empirical Study of Core and Peripheral Developers’ Use of Coding Agents

Technical Papers When: Mon 13 Apr 2026 12:10 - 12:20 People: Shamse Tasnim Cynthia, Joy Krishan Das, Banani Roy

… …

DBSecQA: A Curated Dataset of Developer Discussions on Database Security from Stack Exchange

Data and Tool Showcase Track When: Mon 13 Apr 2026 16:00 - 17:30 People: Md Rakibul Islam, Farha Kamal, MD HUMAUN KABIR, Md Murad Sharif

… revealing that encryption and authentication represent 41% of all discussions … questions (62.93%). The dataset, along with all collection and processing scripts …

The Role of an Empirical Software Engineering Researcher in the Age of Generative AI

Keynotes People: Emerson Murphy-Hill

… token-rich? And why write papers at all when generative AI can both synthesize …

GLiSE: A Prompt-Driven and ML-Powered Tool for Automated Grey Literature Extraction in Software Engineering

Data and Tool Showcase Track When: Mon 13 Apr 2026 16:00 - 17:30 People: Brahim Mahmoudi, Zacharie Chenail-Larcher, Houcine Abdelkader Cherief, Quentin Stiévenart, Naouel Moha, Florent AVELLANEDA

… to their relevance. \textsc{GLiSE} is designed for reproducibility with all

LGTM! Characteristics of Auto-Merged LLM-based Agentic PRs

Mining Challenge When: Mon 13 Apr 2026 14:00 - 15:30 People: Ruben Branco, Paulo Canelas, Catarina Gamboa, Alcides Fonseca

… , and that repositories tend to either auto-merge all or none agentic PRs, with more mature …

Source Code Hotspots: A Diagnostic Method for Quality Issues

Technical Papers When: Tue 14 Apr 2026 11:00 - 11:10 People: Saleha Muzammil, Mughees Ur Rehman, Zoe Kotti, Diomidis Spinellis

… generate 74% of all hotspot edits, suggesting that bot activity is a dominant …

Test Coverage of Code Changes in AI-Generated Pull Requests

Mining Challenge When: Mon 13 Apr 2026 14:00 - 15:30 People: Tales Alves, Leopoldo Teixeira

… compare three authorship types: AI-Only (all commits authored by agents), Coauthor …) appear in over 90% of test files across all groups, with no statistically …

Testing with AI Agents: An Empirical Study of Test Generation Frequency, Quality, and Coverage

Mining Challenge When: Mon 13 Apr 2026 14:00 - 15:30 People: Suzuka Yoshimoto, Shun Fujita, Kosei Horikawa, Daniel Feitosa, Yutaro Kashiwa, Hajimu Iida

… coverage. Our findings reveal that (1) AI authored 16.4% of all commits adding …

OmniCCG: Agnostic Code Clone Genealogy Extractor

Data and Tool Showcase Track When: Mon 13 Apr 2026 16:00 - 17:30 People: Denis Sousa, Matheus Paixao, Thiago Lima, Adriely Silva, Italo Uchoa, Chaiyong Ragkhitwetsagul

… genealogy extractor exists. Furthermore, all extractors proposed …

The Dose Makes the Agent: Therapeutic Index Analysis of AI Coding Contributions

Mining Challenge When: Mon 13 Apr 2026 14:00 - 15:30 People: Giuseppe Destefanis, Ronnie de Souza Santos, Marco Ortu, Mairieli Wessel

… ) and toxicity (review friction). All agents exhibit statistically significant …

Mapping Decentralized Autonomous Organization Governance Across Chains: An Updated, Multi-Platform Dataset

Data and Tool Showcase Track When: Mon 13 Apr 2026 16:00 - 17:30 People: Mashiat Amin Farin, Samer Hassan, Javier Arroyo

… ) communities. The dataset and code are all publicly available and open-licensed …

Mining Type Constructs Using Patterns in AI-Generated Code

Mining Challenge When: Mon 13 Apr 2026 14:00 - 15:30 People: Imgyeong Lee, Tayyib Ul Hassan, Abram Hindle

… . Surprisingly, even with all these issues, Agentic pull requests (PRs) have 1.8 …

Reliability of AI Bots Footprints in GitHub Actions CI/CD Workflows

Mining Challenge When: Mon 13 Apr 2026 14:00 - 15:30 People: Syed Muhammad Ashhar Shah, Sehrish Habib, Muizz Ahmed Hussain, Maryam Abdul Ghafoor, Abdul Ali Bangash

… the GitHub Actions API and analyzed 61,837 runs from 2,355 repositories, all

Learning Compiler Fuzzing Mutators from Historical Bugs

Technical Papers When: Tue 14 Apr 2026 14:40 - 14:50 People: Lingjun Liu, Feiran Qin, Owolabi Legunsen, Marcelo d'Amorim

… . Then, we run IssueMut on these compilers, with all their test inputs as seed …

Analyzing Dependency Distribution Changes Arising from Code Smell Interactions

Technical Papers When: Tue 14 Apr 2026 12:00 - 12:10 People: Zushuai Zhang, Elliott Wen, Ewan Tempero

… cases, and that all code smells are associated with a consistent change direction …

How AI Coding Agents Modify Code: A Large-Scale Study of GitHub Pull Requests

Mining Challenge When: Mon 13 Apr 2026 14:00 - 15:30 People: Daniel Ogenrwot, John Businge

… across all measures. These findings provide a large-scale empirical …

Do AI Agents Really Improve Code Readability?

Mining Challenge When: Mon 13 Apr 2026 14:00 - 15:30 People: Kyogo Horikawa, Kosei Horikawa, Yutaro Kashiwa, Hidetake Uwano, Hajimu Iida

… cover all forms of refactoring, including performance optimization …

Beyond the Prompt: Assessing Domain Knowledge Strategies for High-Dimensional LLM Optimization in Software Engineering

Registered Reports When: Mon 13 Apr 2026 16:45 - 16:50 People: Srinath Srinivasan, Tim Menzies

… that all human studies conducted as part of this study will comply …

Skyt: Prompt Contracts for Software Repeatability in LLM-Assisted Development

Data and Tool Showcase Track When: Mon 13 Apr 2026 16:00 - 17:30 People: Heitor Roriz Filho, Nasser Jazdi, Vicente Lucena

… ), multiple temperatures and contract types, plus scripts to recompute all tables …

Comparing AI Coding Agents: A Task-Stratified Analysis of Pull Request Acceptance

Mining Challenge When: Mon 13 Apr 2026 14:00 - 15:30 People: Giovanni Pinna, Jingzhi Gong, David Williams, Federica Sarro

… high acceptance rates across all nine task categories (59.6%–88.6 …

An Empirical Study on Line-Level Software Defect Prediction

Technical Papers When: Mon 13 Apr 2026 11:30 - 11:40 People: Enci Zhang, Yutong Jiang, Tianmeng Zhang, Haonan Tong

… these models is statistically significant; 2) no model is always superior across all

A Study on Code Clone Lifecycles in Pull Requests Created by AI Agents

Mining Challenge When: Mon 13 Apr 2026 14:00 - 15:30 People: Italo Uchoa, Denis Sousa, Henrique Chuvas, Matheus Paixao, Chaiyong Rakhitwetsagul, Thiago Lima

… require consistent updates across all related clones, negatively affecting software …

Beyond Bug Fixes: An Empirical Investigation of Post-Merge Code Quality Issues in Agent-Generated Pull Requests

Mining Challenge When: Mon 13 Apr 2026 14:00 - 15:30 People: Shamse Tasnim Cynthia, Al Muttakin, Banani Roy

… are primarily driven by larger PRs. Across all agents, code smells dominate, particularly …

Model See, Model Do? Exposure-Aware Evaluation of Bug-vs-Fix Preference in Code LLMs

Technical Papers When: Mon 13 Apr 2026 11:20 - 11:30 People: Ali Al-Kaswan, Claudio Spiess, Prem Devanbu, Arie van Deursen, Maliheh Izadi

… prefer the fixed code across all conditions, indicating a stable bias toward …

An Empirical Study of Policy as Code: Adoption, Purpose, and Maintenance

Technical Papers When: Tue 14 Apr 2026 14:50 - 15:00 People: Ruben Opdebeeck, Mahmoud Alfadel, Akond Rahman, Yutaro Kashiwa, João F. Ferreira, Raula Gaikovina Kula, Coen De Roover

… find that PaC is introduced throughout all phases of repository lifecycles, often …