On the Shape of Brainscores for Large Language Models (LLMs)
This work addresses the need for interpretability in interdisciplinary AI-neuroscience research by providing insights into a novel metric, though it appears incremental as it builds on existing brainscore concepts without major methodological breakthroughs.
The study tackled the problem of interpreting the novel 'Brainscore' metric for evaluating functional similarity between Large Language Models (LLMs) and human brain systems by constructing topological features from fMRI data of 190 subjects and 39 LLMs, and training 36 Linear Regression Models to identify reliable features. The result revealed distinctive feature combinations that help interpret brainscores across brain regions and hemispheres, contributing to interpretable machine learning.
With the rise of Large Language Models (LLMs), the novel metric "Brainscore" emerged as a means to evaluate the functional similarity between LLMs and human brain/neural systems. Our efforts were dedicated to mining the meaning of the novel score by constructing topological features derived from both human fMRI data involving 190 subjects, and 39 LLMs plus their untrained counterparts. Subsequently, we trained 36 Linear Regression Models and conducted thorough statistical analyses to discern reliable and valid features from our constructed ones. Our findings reveal distinctive feature combinations conducive to interpreting existing brainscores across various brain regions of interest (ROIs) and hemispheres, thereby significantly contributing to advancing interpretable machine learning (iML) studies. The study is enriched by our further discussions and analyses concerning existing brainscores. To our knowledge, this study represents the first attempt to comprehend the novel metric brainscore within this interdisciplinary domain.