About — Ivan Ilin

Research with a path to practice

Ivan Ilin is a machine learning researcher and PhD candidate in Computer Science at KAUST. His work focuses on efficient optimization and compression methods for large language models, including pruning, sparse fine-tuning, quantization, and pipeline parallelism. He also develops open-source tools that apply machine learning to practical workflows.

Current focus

I am a PhD candidate in Computer Science at KAUST. My current work examines how large language models can be trained, adapted, and deployed with less computation and memory. The recurring ideas are sparsity, low precision, second-order optimization, and distributed model structure.

I am also interested in translating research and machine-learning systems into practical tools. VoiceCut is one example: an open-source project built around a repetitive problem in narration editing.

For research details, see the research overview, the publication list, or the publicproject pages.