Wayne Xin Zhao
The synthesis side of LLM research — what it takes to read every paper of the moment and produce something other researchers can navigate.
Zhao led the writing of "A Survey of Large Language Models" — the 100+ page synthesis paper that became, for a stretch of 2023, the most-cited reference for what the LLM field actually contained. The work behind that kind of paper is less visible than the methodological papers that produced the surveyed work: someone has to read everything, identify which claims actually survive scrutiny when laid next to each other, and structure the result so it remains useful when the field moves another six months. His longer research line in information retrieval and recommender systems gives the survey work its underlying methodological discipline — the same instincts that catch failure modes in recommender-eval transfer cleanly to catching them in LLM-eval claims.