In-Context Semi-Supervised Learning

TL;DR

This paper introduces in-context semi-supervised learning (IC-SSL), demonstrating how Transformers can leverage unlabeled data alongside few labels to improve representation learning and prediction accuracy in low-label scenarios.

Contribution

It presents the concept of IC-SSL, showing how Transformers utilize unlabeled context for robust, context-dependent representations, advancing understanding of in-context learning with limited labels.

Findings

Transformers effectively leverage unlabeled context for learning.

IC-SSL improves performance in low-label regimes.

Provides foundational insights into unlabeled data exploitation.

Abstract

There has been significant recent interest in understanding the capacity of Transformers for in-context learning (ICL), yet most theory focuses on supervised settings with explicitly labeled pairs. In practice, Transformers often perform well even when labels are sparse or absent, suggesting crucial structure within unlabeled contextual demonstrations. We introduce and study in-context semi-supervised learning (IC-SSL), where a small set of labeled examples is accompanied by many unlabeled points, and show that Transformers can leverage the unlabeled context to learn a robust, context-dependent representation. This representation enables accurate predictions and markedly improves performance in low-label regimes, offering foundational insights into how Transformers exploit unlabeled context for representation learning within the ICL framework.