Saved in:
Bibliographic Details
Main Authors: Fan, Jiashuo, Rosu, Paul, Wang, Aaron T., Li, Zeyu Michael, Carin, Lawrence, Cheng, Xiang
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.15934
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910005268054016
author Fan, Jiashuo
Rosu, Paul
Wang, Aaron T.
Li, Zeyu Michael
Carin, Lawrence
Cheng, Xiang
author_facet Fan, Jiashuo
Rosu, Paul
Wang, Aaron T.
Li, Zeyu Michael
Carin, Lawrence
Cheng, Xiang
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2512_15934
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle In-Context Semi-Supervised Learning
Fan, Jiashuo
Rosu, Paul
Wang, Aaron T.
Li, Zeyu Michael
Carin, Lawrence
Cheng, Xiang
Machine Learning
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.
title In-Context Semi-Supervised Learning
topic Machine Learning
url https://arxiv.org/abs/2512.15934