Spontaneous Speech Variables for Evaluating LLMs Cognitive Plausibility
This work addresses the need for better cognitive evaluation of LLMs, particularly for researchers in NLP and cognitive science, but it is incremental as it extends existing evaluation methods to new variables.
The paper tackled the problem of evaluating large language models from a cognitive perspective by using spontaneous speech corpora to derive production variables like speech reductions and prosodic prominences, and found that after fine-tuning, models predicted these variables well above baselines, with spoken genre training data yielding more accurate predictions than written genres.
The achievements of Large Language Models in Natural Language Processing, especially for high-resource languages, call for a better understanding of their characteristics from a cognitive perspective. Researchers have attempted to evaluate artificial models by testing their ability to predict behavioral (e.g., eye-tracking fixations) and physiological (e.g., brain responses) variables during language processing (e.g., reading/listening). In this paper, we propose using spontaneous speech corpora to derive production variables (speech reductions, prosodic prominences) and applying them in a similar fashion. More precisely, we extract. We then test models trained with a standard procedure on different pretraining datasets (written, spoken, and mixed genres) for their ability to predict these two variables. Our results show that, after some fine-tuning, the models can predict these production variables well above baselines. We also observe that spoken genre training data provides more accurate predictions than written genres. These results contribute to the broader effort of using high-quality speech corpora as benchmarks for LLMs.