To Words and Beyond: Probing Large Language Models for Sentence-Level Psycholinguistic Norms of Memorability and Reading Times
This work addresses the challenge of estimating sentence-level cognitive features for researchers in psycholinguistics and NLP, but it is incremental as it extends existing methods to new norms.
The authors tackled the problem of predicting sentence-level psycholinguistic norms like memorability and reading times using large language models, finding that fine-tuned models achieved correlations with human judgments and outperformed baseline predictors, while zero-shot and few-shot prompting performed poorly.
Large Language Models (LLMs) have recently been shown to produce estimates of psycholinguistic norms, such as valence, arousal, or concreteness, for words and multiword expressions, that correlate with human judgments. These estimates are obtained by prompting an LLM, in zero-shot fashion, with a question similar to those used in human studies. Meanwhile, for other norms such as lexical decision time or age of acquisition, LLMs require supervised fine-tuning to obtain results that align with ground-truth values. In this paper, we extend this approach to the previously unstudied features of sentence memorability and reading times, which involve the relationship between multiple words in a sentence-level context. Our results show that via fine-tuning, models can provide estimates that correlate with human-derived norms and exceed the predictive power of interpretable baseline predictors, demonstrating that LLMs contain useful information about sentence-level features. At the same time, our results show very mixed zero-shot and few-shot performance, providing further evidence that care is needed when using LLM-prompting as a proxy for human cognitive measures.