**Overview**

Research Internships at Microsoft provide a dynamic environment for research careers with a network of world-class research labs led by globally-recognized scientists and engineers, who pursue innovation in a range of scientific and technical disciplines to help solve complex challenges in diverse fields, including computing, healthcare, economics, and the environment.

Improving efficiency of Large Language Models (LLMs) is critical to deploying large models, and training (coupled with other techniques such as quantization) is a promising tool for improving efficiency.

This Research Internship will design training algorithms and apply them to improving the quality/efficiency trade-offs of large language models, with a focus on resource-constrained environments.

Possible directions of investigation include: designing new algorithms for quantized model fine-tuning; leveraging training to improve the token efficiency of reasoning models; pr...