arXiv:2611.13926v1 Announce Type: new
Abstract: Explainable AI (XAI) has expanded into a broad and impactful research area, but the fundamental question of how best to evaluate explanation techniques has proven frustratingly elusive. Methodologies typically revolve around either proxy metrics which tell us little about the utility of an explanation, or human evaluation which---while generally considered the gold standard---is expensive, difficult to scale, and challenging to reproduce. In this paper, we investigate if Multimodal Large Language Models (MLLMs) can act as surrogate human participants to automate the evaluation of XAI. To do this, we first propose a ``purpose-grounded'' evaluation framework with metrics applicable to both MLLMs and human users alike. Then, we offer some conceptual insights as to why and when we can trust these models to automate this process. Lastly, we conduct extensive empirical testing across five purpose-grounded areas such as model improvement and fourteen popular MLLMs. Results indicate that all models trend positively, but OpenAI and Qwen models generally show higher alignment compared to other variants, which would be the recommended choices in practice. We expect our insights not to replace user testing, but rather to help XAI researchers complement their current evaluation methodologies and move the field away from heavy reliance on proxy metrics for measuring explanation utility.
arxiv
MLLMs as Human Surrogates for Explainable AI Evaluation
This paper investigates whether multimodal large language models can act as surrogate human participants to automate the evaluation of explainable AI techniques. It proposes a purpose-grounded evaluation framework with metrics applicable to both MLLMs and human users, and offers conceptual arguments for when such automation can be trusted. Empirical testing covers five purpose-grounded areas and fourteen popular MLLMs, finding that all models trend positively while OpenAI and Qwen variants generally show higher alignment. The authors present the approach as a complement to, not a replacement for, user testing.
It offers an empirical framework for scaling XAI evaluation beyond proxy metrics by testing MLLMs as human surrogates.
Comments & Questions (0)
No comments yet
Be the first to comment!