Everything Everywhere All at Once: LLMs can In-Context Learn Multiple Tasks in Superposition

Zheyang Xiong; Ziyang Cai; John Cooper; Albert Ge; Vasilis Papageorgiou; Zack Sifakis; Angeliki Giannou; Ziqian Lin; Liu Yang; Saurabh Agarwal; Grigorios Chrysos; Samet Oymak; Kangwook Lee; Dimitris Papailiopoulos

2025 ICML ICML 2025

Everything Everywhere All at Once: LLMs can In-Context Learn Multiple Tasks in Superposition

Abstract

No abstract available. No abstract was captured for this paper when its proceedings were scraped - most often an older edition published without one online. How the corpus is built.

Authors

Zheyang Xiong, Ziyang Cai, John Cooper, Albert Ge, Vasilis Papageorgiou, Zack Sifakis, Angeliki Giannou, Ziqian Lin, Liu Yang, Saurabh Agarwal, Grigorios Chrysos, Samet Oymak, Kangwook Lee, Dimitris Papailiopoulos

Topics

Artificial Intelligence > Core AI > Foundation Models Artificial Intelligence > Core AI > Multimodal Learning Artificial Intelligence > Core AI > Large Language Models

Keywords

in-context learning transformer output distribution llms simulator task superposition

Download PDF