July 2 (Thu) – July 4 (Sat), 2026
Fostering the Next Generation of AI Researchers
Invited Talks

Gianmario Voria
Towards Fair and Trustworthy AI Assistants

Ali Arabat
Characterizing Topics in Agent Context Files: Prevalence, Characteristics, and Insights
Program
July 2 (Thu) – July 4 (Sat), 2026
Program Overview
Session-level program.
July 2 (Thu) Room #612 · English Session
| 13:00–13:15 | Opening |
| 13:15–14:15 | Session 1 |
| 14:15–14:45 | Break |
| 14:45–15:45 | Session 2 |
| 15:45–16:15 | Break |
| 16:15–17:15 | Session 3 |
| 17:15–18:30 | Break |
| 18:30–21:00 | Dinner |
July 3 (Fri) Room #612 · English Session
| 9:00–10:00 | Session 4 (Invited Talks) |
| 10:00–10:30 | Break |
| 10:30–11:15 | Session 5 |
| 11:15–11:45 | Break |
| 11:45–12:30 | Session 6 |
| 12:30–14:00 | Break |
| 14:00–19:00 | Networking Walk in Miyajima |
July 4 (Sat) Room #612 · Japanese Session
| 9:00–10:30 | Session 7 |
| 10:30–11:00 | Break |
| 11:00–12:00 | Tutorial |
| 12:00–12:15 | Closing |
Detailed Program
Each presentation slot is 15 minutes.
July 2 (Thu) Room #612 · English Session
| Time | Speaker | Title |
|---|---|---|
| 13:00 | Opening | |
| 13:15 | Renya Ishida | An Empirical Study of Speedrun Community for Game Software Development |
| 13:30 | Sila Sonpee | PerfSATD: An Empirical Study on Performance-related SATD |
| 13:45 | Hidetake Tanaka | Introducing Spec-Driven Development into Software Development PBL |
| 14:00 | Haruka Tokumasu | Discovery and Application of Programming-Language-Specific Neurons |
| 14:15–14:45 | Break | |
| 14:45 | Kou Hamada | AI Policies in Open Source Software Projects: Governance and Impact |
| 15:00 | Kaito Shigyo | Evaluating Coding Agents for Refactoring Under Project-Specific Conventions |
| 15:15 | Xia Sisi | Do Agent Context Files Provide Security Guardrails? An Empirical Study of Agentic AI Software Repositories |
| 15:30 | Shakizada Yagut | The Roles of Humans in AI-Centric Development: A Large-Scale Study of Multi-Agent Pull Request Workflows |
| 15:45–16:15 | Break | |
| 16:15 | Tatsuya Shirai | Large-Scale Empirical Analysis of Continuous Fuzzing: Insights from 1 Million Fuzzing Sessions |
| 16:30 | Kuan Kiet Ng | An Empirical Study of Reproducible OSS-Fuzz Build Failure |
| 16:45 | Kan Watanabe | On the Impact of Mutation Testing on Just-In-Time Defect Prediction |
| 17:00 | Akihiro Saeki | Information Retrieval based Fault Localization with LLM |
| 17:15–18:30 | Break | |
| 18:30–21:00 | Dinner (map) | |
July 3 (Fri) Room #612 · English Session
| Time | Speaker | Title |
|---|---|---|
| 9:00–9:30 | Ali Arabat | Invited talk: Characterizing Topics in Agent Context Files: Prevalence, Characteristics, and Insights |
| 9:30–10:00 | Gianmario Voria | Invited talk: Towards Fair and Trustworthy AI Assistants |
| 10:00–10:30 | Break | |
| 10:30 | Daiki Kazusaki | How the Inside of an LLM Grows as It Learns Code2Code Tasks: Understanding and Application |
| 10:45 | Warren Polandra | Understanding Agentic Specifications and Generated Code |
| 11:00 | Kosei Horikawa | Can We Distinguish Between AI Agent Commits and Human Commits? Characterization, Detection, and Limits |
| 11:15–11:45 | Break | |
| 11:45 | Pongchai Jaisri | How Software Repository are Mined: A Preliminary Systematic Literature Review |
| 12:00 | Pakorn Laohakanniyom | Your Latest is Outdated: An Empirical Study of Latest Tags in Dockerfiles |
| 12:15 | Nigar Alizada | PyClone-Stream: An Automated and Self-growing Python Semantic Clone Dataset |
| 12:30–14:00 | Break | |
| 14:00–19:00 | Networking Walk in Miyajima | |
July 4 (Sat) Room #612 · Japanese Session
| Time | Speaker | Title |
|---|---|---|
| 9:00 | Ryotaro Yamamoto | Do Coding Agents Really Follow Their Context Files? An Empirical Behavioral Evaluation |
| 9:15 | Ryosuke Ono | How Test Code Quality Affects Bug-Fixing Cost of Autonomous Software Agents |
| 9:30 | Sota Onuma | When Tests Fail to Fail: Detecting Silent Horrors in Test Code Using LLMs |
| 9:45 | Hirohito Ota | An Investigation into the Effect of Method Ordering on LLM Code Comprehension |
| 10:00 | Kota Kojiro | Identification of Task-Specific Neurons in Software Engineering |
| 10:15 | Yuki Ifuku | Predicting the Acceptance or Rejection of Proposals in Go Language Development with LLM |
| 10:30–11:00 | Break | |
| 11:00–11:30 | Masanari Kondo | Tutorial (To appear) |
| 11:30–12:00 | Yutaro Kashiwa | Tutorial (To appear) |
| 12:00 | Closing |
Organizers

Yutaro Kashiwa
NAIST

Masanari Kondo
Kyushu University
Venue
RCC Culture Center <Hashimoto-cho>, Hiroshima
(HP: https://www.rccbc.co.jp/kaigi/rccbc/)
(Map: https://maps.app.goo.gl/LghTLiFB2wotKoh56)