Skip to content
conftrace_
2025 ICML ICML 2025

Chameleon: A Flexible Data-mixing Framework for Language Model Pretraining and Finetuning

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

Wanyun Xie, Francesco Tonin, Volkan Cevher

Topics

Machine Learning > Core Methods > Representation Learning Machine Learning > Application Areas > Domain Adaptation Artificial Intelligence > Core AI > Large Language Models

Keywords

embedding space leverage score few-shot reasoning data mixing test perplexity domain affinity matrix
Download PDF