Jiahui Yu

One of the more useful people to study for the Gemini era because his work spans both the text-core of multimodal frontier models and the optimization tricks that make those systems cheaper and more stable to train.

A high-signal researcher for understanding the modern scaling playbook, especially around compute-optimal training, retrieval-augmented language models, and the text side of Gemini-era multimodal systems.

One of the clearest multimodal researchers to track if you want to understand how frontier labs turned vision-language work from narrow benchmarks into general-purpose model capability.

Important for understanding how multilingual NLP, translation, and multimodal reasoning meet inside production-scale frontier systems rather than staying separate research tracks.

A good researcher to follow for the infrastructure side of frontier language models, especially mixture-of-experts scaling, instruction tuning, and the data systems that make very large models usable.

Worth tracking for the data side of multimodal frontier models, where the quality and shape of training mixtures strongly determine what large systems can actually do.