The Mystery of In-Context Learning: A Comprehensive Survey on Interpretation and Analysis

TL;DR

ICL enables LLMs to perform tasks using demonstrations without re-training, raising questions about underlying mechanisms and risks. The paper surveys theoretical (mechanistic, regression, gradient-descent/meta-optimization, Bayesian) and empirical (pre-training data/model, demonstration order, input-label mappings) analyses. It highlights contradictory findings, challenging evaluation and demonstrating need for causal, synthetic, and safety-focused research. The authors provide a framework and a resources repository to advance interpretation of ICL.

Abstract

Understanding in-context learning (ICL) capability that enables large language models (LLMs) to excel in proficiency through demonstration examples is of utmost importance. This importance stems not only from the better utilization of this capability across various tasks, but also from the proactive identification and mitigation of potential risks, including concerns regarding truthfulness, bias, and toxicity, that may arise alongside the capability. In this paper, we present a thorough survey on the interpretation and analysis of in-context learning. First, we provide a concise introduction to the background and definition of in-context learning. Then, we give an overview of advancements from two perspectives: 1) a theoretical perspective, emphasizing studies on mechanistic interpretability and delving into the mathematical foundations behind ICL; and 2) an empirical perspective, concerning studies that empirically analyze factors associated with ICL. We conclude by highlighting the challenges encountered and suggesting potential avenues for future research. We believe that our work establishes the basis for further exploration into the interpretation of in-context learning. Additionally, we have created a repository containing the resources referenced in our survey.

Figures (1)

• Figure 1: Illustration of In-context Learning.