Colloquium with Tal Linzen, NYU
Title: To model human linguistic cognition, make LLMs less superhuman
To model human linguistic cognition, make LLMs less superhuman
When people listen to or read a sentence, they actively make predictions about upcoming words: words that are less predictable are generally read more slowly than predictable ones. The success of large language models (LLMs), which, like humans, make predictions about upcoming words, has motivated exploring the use of these models as cognitive models of human linguistic prediction. Surprisingly, in the last few years, as language models have become better at predicting the next word, their ability to predict human reading behavior has declined. This is because LLMs are able to predict upcoming words much better than people can, leading them to predict lower processing difficulty in reading than observed in human experiments; in other words, mainstream LLMs are ‘superhuman’ as models of language comprehension. LLMs’ superhumanness is primarily driven by two factors: compared to humans, LLMs have much stronger long-term memory for facts and training examples, and they have much better short-term memory for previous words in the text. In this talk, I'll survey some of the work from my group and others that supports these hypotheses, and will outline some ongoing and future work possible directions for creating more human-like LLMs.