Is In-Context Learning a Type of Error-Driven Learning? Evidence from the Inverse Frequency Effect in Structural Priming

TL;DR

This paper investigates whether in-context learning (ICL) in large language models (LLMs) functions as error-driven learning by examining the inverse frequency effect (IFE), a phenomenon linked to error-driven mechanisms in psycholinguistics.

Contribution

The study introduces a novel diagnostic approach using IFE to determine if ICL is error-driven, providing evidence that LLMs exhibit IFE similar to humans, especially in larger models.

Findings

LLMs display the inverse frequency effect in structural priming.

The IFE effect is stronger in larger language models.

Results support that ICL involves error-driven learning mechanisms.

Abstract

Large language models (LLMs) have shown the emergent capability of in-context learning (ICL). One line of research has claimed that ICL is functionally equivalent to gradient descent, a type of error-driven learning mechanism. In this paper, we introduce a new way of diagnosing whether ICL is functionally performing error-driven learning. Our approach is based on the inverse frequency effect (IFE) -- a phenomenon in which an agent's behavior is influenced to a greater degree when presented with improbable examples as compared to more likely ones. The IFE has previously been identified in psycholinguistics where humans exhibit the IFE in the context of structural priming (the tendency for people to produce sentence structures they have encountered recently). In that context, the IFE has been used as evidence that human structural priming must involve error-driven learning mechanisms. In our experiments, we simulated structural priming with ICL and found that LLMs indeed display the IFE, with the effect being stronger in larger models. We conclude that at least in the case we studied, ICL is indeed a type of error-driven learning, supporting the hypothesis that an error signal is implicitly computed in the forward pass during ICL. Our results suggest that both humans and LLMs make use of error-driven processing mechanisms in on-line processing.