Dongyang Fan

Hi! Thanks for stopping by :) I'm a passionate PhD student at Machine Learning and Optimization Lab at EPFL, supervised by Prof. Martin Jaggi. My name is pronounced as Don-Young.

My research interests include:

Efficient Language Modeling
- Mixture-of-Experts architectures
- Decentralized training methods
- Accelerating LLM pretraining through metadata conditioning
Responsible Language Modeling
- Quantifying the "Data Compliance Gap"
- Designing compensation frameworks for data contributors
- Understanding and mitigating model hallucinations

I am also happy to branch out my research. If you want to reach out, do not hesitate to drop me an email!

Email / Google Scholar / Twitter / Github / LinkedIn

Research

	TiMoE: Time-Aware Mixture of Language Experts Robin Faro, Dongyang Fan, Tamar Alphaidze, Martin Jaggi XTempLLMs workshop @ COLM (oral), 2025 arXiv
	URLs Help, Topics Guide: Understanding Metadata Utility in LLM Training Dongyang Fan, Vinko Sabolčec, Martin Jaggi, preprint, 2025 arXiv
	Can Performant LLMs Be Ethical? Quantifying the Impact of Web Crawling Opt-Outs Dongyang Fan, Vinko Sabolčec, Matin Ansaripour, Ayush Kumar Tarun, Martin Jaggi, Antoine Bosselut, Imanol Schlag Conference on Language Modeling (COLM), 2025 arXiv
	From Fairness to Truthfulness: Rethinking Data Valuation Design Dongyang Fan, Tyler J. Rotello, Sai Praneeth Karimireddy ICLR Workshop Data Problems, 2025 arXiv
	On-Device Collaborative Language Modeling via a Mixture of Generalists and Specialists Dongyang Fan, Bettina Messmer, Nikita Doikov, Martin Jaggi International Conference on Machine Learning (ICML), 2025 codes / arXiv
	Towards an empirical understanding of MoE design choices Dongyang Fan, Bettina Messmer, Martin Jaggi ICLR ME-FoMo Workshop, 2024 arXiv
	Personalized Collaborative Fine-Tuning for On-Device Large Language Models Nicolas Wagner, Dongyang Fan, Martin Jaggi Conference on Language Modeling (COLM), 2024 codes / arXiv
	Ghost Noise for Regularizing Deep Neural Networks Atli Kosson, Dongyang Fan, Martin Jaggi Association for the Advancement of Artificial Intelligence (AAAI), 2024 arXiv
	Collaborative Learning via Prediction Consensus Dongyang Fan, Celestine Mendler-Dünner, Martin Jaggi Conference on Neural Information Processing Systems (NeurIPS), 2023 codes / arXiv / poster

Miscellaneous

In general I like arts and cultural stuff. I am also an outdoorsy person and I do hiking skiing and sailing.

I paint from my hiking trips. For example...

Source codes of the website are from here.