The AI So-Go-Chi Center is pleased to announce that the following paper, co-authored by Assistant Professor Watanee Jearanaiwongkul and Associate Professor Teeradaj Racharak, has been accepted for presentation at the 25th International Conference on Intelligent Software Methodologies, Tools, and Techniques (SOMET 2026). The conference will be held in Kuala Lumpur, Malaysia, from September 9 to 11, 2026. SOMET is an international conference that brings together researchers and practitioners to present advances in intelligent software engineering, software methodologies, and AI-driven software technologies.

Accepted Paper

Watanee Jearanaiwongkul, Teeradaj Racharak, Chanidapha Soipool, Benjawan Chompootepa,
Metamorphic Testing for Assessing Behavioral Consistency of Open-source Large Language Models, Proceedings of the 25th International Conference on Intelligent Software Methodologies, Tools, and Techniques (SOMET), Malaysia, September 9–11, 2026.

This paper investigates the behavioral consistency of open-source large language models (LLMs) using Metamorphic Testing (MT), a testing methodology that enables the evaluation of AI systems without requiring ground-truth labels. The proposed approach assesses whether LLMs produce consistent and reliable outputs under semantically equivalent input transformations. By systematically analyzing behavioral changes across multiple open-source LLMs, the study provides valuable insights into their robustness, reliability, and trustworthiness, contributing to the development of more dependable AI systems.

Congratulations to the authors on this achievement!