Our research on psychometric evaluation of LLMs was accepted to the CHI 2025 HEAL workshop. This work explores new methods for understanding user experiences with large language models through validated psychological scales.
The paper presents our framework for applying psychometric principles to LLM evaluation, offering a more nuanced approach to understanding how users perceive and interact with these systems.
Looking forward to discussing this work with the HEAL workshop community and exploring future directions for human-centered evaluation of AI systems.