Augmenting Interpretable Models with LLMs during Training
This addresses the problem of deploying AI in high-stakes and compute-limited settings by providing interpretable and efficient models, though it is incremental as it builds on existing interpretable methods.
The authors tackled the need for interpretability and efficiency in high-stakes domains by proposing Augmented Interpretable Models (Aug-imodels), a framework that leverages LLMs during training to build efficient and transparent models, achieving over 1,000x speed/memory improvement in inference and outperforming larger models like GPT-J with 10,000x fewer parameters.
Recent large language models (LLMs) have demonstrated remarkable prediction performance for a growing array of tasks. However, their proliferation into high-stakes domains (e.g. medicine) and compute-limited settings has created a burgeoning need for interpretability and efficiency. We address this need by proposing Augmented Interpretable Models (Aug-imodels), a framework for leveraging the knowledge learned by LLMs to build extremely efficient and interpretable models. Aug-imodels use LLMs during fitting but not during inference, allowing complete transparency and often a speed/memory improvement of greater than 1,000x for inference compared to LLMs. We explore two instantiations of Aug-imodels in natural-language processing: (i) Aug-GAM, which augments a generalized additive model with decoupled embeddings from an LLM and (ii) Aug-Tree, which augments a decision tree with LLM feature expansions. Across a variety of text-classification datasets, both outperform their non-augmented counterparts. Aug-GAM can even outperform much larger models (e.g. a 6-billion parameter GPT-J model), despite having 10,000x fewer parameters and being fully transparent. We further explore Aug-imodels in a natural-language fMRI study, where they generate interesting interpretations from scientific data. All code for using Aug-imodels and reproducing results is made available on Github.