SSBSE 2025
Sun 16 Nov 2025 Seoul, South Korea
co-located with ASE 2025
Sun 16 Nov 2025 14:10 - 14:30 at Grand Hall 4 - SSBSE Challenge

Coding agents powered by Large Language Models (LLMs) face critical sustainability and scalability challenges in industrial deployment, often incurring costs that may exceed optimization benefits. We introduce GA4GC, the first framework to optimize coding agent runtime (greener agent) and code performance (greener code) trade-offs by discovering Pareto-optimal agent hyperparameters and prompt templates. Evaluation on the SWE-Perf benchmark demonstrates up to 135-fold hypervolume improvement, reducing agent runtime by 37.7% while improving correctness. Baseline comparisons and influence analysis confirm the effectiveness of GA4GC, identify temperature as the most influential hyperparameter, and provide actionable strategies to balance agent and code sustainability in industrial deployment.

Sun 16 Nov

Displayed time zone: Seoul change

14:00 - 15:30
SSBSE ChallengeSSBSE Challenge at Grand Hall 4
14:00
10m
Talk
Challenge Overview
SSBSE Challenge

14:10
20m
Talk
GA4GC: Greener Agent for Greener Code via Multi-Objective Configuration Optimization
SSBSE Challenge
Jingzhi Gong University of Leeds, Yixin Bian Harbin Normal University, Luis de la Cal Universidad Politécnica de Madrid, Giovanni Pinna University of Trieste, Anisha Uteem King's College London, David Williams University College London, Mar Zamorano López University College London, Karine Even-Mendoza King’s College London, William Langdon University College London, Hector Menendez King’s College London, Federica Sarro University College London
Pre-print
14:30
20m
Talk
GreenMalloc: Allocator Optimisation for Industrial Workloads
SSBSE Challenge
Aidan Dakhama King's College London, William Langdon University College London, Hector Menendez King’s College London, Karine Even-Mendoza King’s College London
Pre-print
14:50
20m
Talk
Fuzz Smarter, Not Harder: Towards Greener Fuzzing with GreenAFL
SSBSE Challenge
Ayse Irmak Ercevik King's College London, Aidan Dakhama King's College London, Melane Navaratnarajah King's College London, Yazhuo Cao King's College London, Leo Fernandes Federal Institute of Alagoas (IFAL)
Pre-print File Attached
15:10
20m
Talk
HotCat: Green and Effective Feature Selection for HotFix Bug Taxonomy
SSBSE Challenge
Luis de la Cal Universidad Politécnica de Madrid, Yazhuo Cao King's College London, Ayse Irmak Ercevik King's College London, Giovanni Pinna University of Trieste, Lukas Twist King's College London, David Williams University College London, Karine Even-Mendoza King’s College London, William Langdon University College London, Hector Menendez King’s College London, Federica Sarro University College London
Pre-print