AgenticSE

Program & Accepted Papers (AgenticSE’25)

Program

Time Session Title Authors
8:45–9:00 Opening Welcome and introduction  
9:00–10:00 Session 1 Keynote: Building Jules, Google's first external coding agent Alexander Mossin(Google); Mehadi Hassen (Google)
10:00–10:30 Coffee Break  
10:30–10:55 Session 2 LLMs in Debate: Does Arguing Make Them Better at Detecting Metamorphic Relations? (long paper) Dibyendu Brinto Bose, Yoseph Berhanu Alebachew, Chris Brown
10:55–11:20 Session 2 A 3-Layer Agentic Model for Nonfunctional Requirements in Software Engineering (long paper) Ehsan Zabardast, Tiago Vieira, Tony Gorschek
11:20–11:35 Session 2 Transforming Natural Language into Formal Specifications (talk-only) Kuangxiangzi Liu, Alexander Liggesmeyer, Dhiman Chakraborty, Andreas Zeller
11:35–11:50 Session 2 PRIMA: Enabling User Agency and Control in Mobile GUI Agent Autonomy (talk-only) Ching-Ting Lin, Zhi-Hong Ye, Yung-Ju Chang
11:50–12:00 Session 2 Buffer  
12:00–14:00 Lunch Break  
14:00–15:00 Session 3 Keynote: Trae Agent: SOTA Open-source AI Coding Agent for SWE-bench Chao Peng (ByteDance)
15:00–15:25 Session 3 Leveraging Large Language Models for Cybersecurity Risk Assessment — A Case from Forestry Cyber-Physical Systems (long paper) Fikret Mert Gültekin, Oscar Lilja, Ranim Khojah, Rebekka Wohlrab, Marvin Damschen, Mazen Mohamad
15:25–15:30 Session 3 Buffer  
15:30–16:00 Coffee Break  
16:00–16:15 Session 4 AgentGuard: Runtime Verification of AI Agents (short paper) Roham Koohestani
16:15–16:30 Session 4 Bridging LLM Planning Agents and Formal Methods: A Case Study in Plan Verification (short paper) Keshav Ramani, Vali Tawosi, Salwa Alamir, Daniel Borrajo
16:30–16:55 Session 4 The Last Dependency Crusade: Solving Python Dependency Conflicts with LLMs (long paper) Antony Bartlett, Cynthia C. S. Liem, Annibale Panichella
16:55–17:00 Session 4 Buffer  
17:00–17:15 Closing Wrap-up, acknowledgments, and discussion  

Accepted Papers

Bridging LLM Planning Agents and Formal Methods: A Case Study in Plan Verification

Keshav Ramani (J.P. Morgan AI Research), Vali Tawosi (JP Morgan AI Research), Salwa Alamir (J.P. Morgan AI Research), Daniel Borrajo (J.P. Morgan AI Research)


PRIMA: Enabling User Agency and Control in Mobile GUI Agent Autonomy

Ching-Ting Lin (National Yang Ming Chiao Tung University), Zhi-Hong Ye (National Yang Ming Chiao Tung University), Yung-Ju Chang (National Yang Ming Chiao Tung University)


LLMs in Debate: Does Arguing Make Them Better at Detecting Metamorphic Relations?

Dibyendu Brinto Bose (Virginia Tech), Yoseph Berhanu Alebachew (Virginia Tech), Chris Brown (Virginia Tech)


A 3-Layer Agentic Model for Nonfunctional Requirements in Software Engineering

Ehsan Zabardast (Blekinge Institute of Technology), Tiago Vieira (Independent Researcher), Tony Gorschek (Blekinge Institute of Technology)


Leveraging Large Language Models for Cybersecurity Risk Assessment — A Case from Forestry Cyber-Physical Systems

Fikret Mert Gültekin (Chalmers University of Technology and University of Gothenburg), Oscar Lilja (Chalmers University of Technology and University of Gothenburg), Ranim Khojah (Chalmers University of Technology and University of Gothenburg), Rebekka Wohlrab (Chalmers University of Technology and University of Gothenburg | Carnegie Mellon University), Marvin Damschen (RISE Research Institutes of Sweden), Mazen Mohamad (RISE Research Institutes of Sweden | Chalmers University of Technology and University of Gothenburg)


Transforming Natural Language into Formal Specifications

Kuangxiangzi Liu (Volkswagen AG / Saarland University), Alexander Liggesmeyer (CISPA Helmholtz Center for Information Security), Dhiman Chakraborty (Volkswagen AG), Andreas Zeller (CISPA Helmholtz Center for Information Security)


The Last Dependency Crusade: Solving Python Dependency Conflicts with LLMs

Antony Bartlett (Delft University of Technology), Cynthia C. S. Liem (Delft University of Technology), Annibale Panichella (Delft University of Technology)


AgentGuard: Runtime Verification of AI Agents

Roham Koohestani (JetBrains Research)