Meta tags:
description=Koven Yu is a Ph.D. student at the Computer Science department of Stanford University.;

Headings (most frequently used words):

-

Text of the page (most frequently used words):
fellowship (7), physical (6), stanford (6), research (5), world (5), models (5), the (5), and (5), finalist (4), koven (4), show (3), graduate (3), 2023 (3), with (3), hong (3), xing (3), topic (2), dynamics (2), simulation (2), scene (2), understanding (2), current (2), topics (2), publications (2), all (2), selected (2), engineering (2), phd (2), 2025 (2), how (2), actions (2), generative (2), learning (2), work (2), physics (2), grounded (2), google (2), university (2), computer (2), science (2), indicates, equal, contributions, person, identification, past, date, school, soe, 2020, qualcomm, innovation, 2021, nvidia, 2022, meta, siggraph, asia, best, paper, award, roblox, 2024, meshy, jane, street, awards, study, recreating, simulating, responds, this, requires, developing, next, generation, systems, beyond, simply, scaling, which, lacks, interactability, via, methodology, integrating, modeling, focuses, methodological, principles, apply, them, broader, domains, such, robotics, analysis, earned, department, advised, spent, two, wonderful, summers, working, prior, visited, manmohan, chandraker, san, diego, bill, freeman, noah, snavely, jiajun, vision, lab, twitter, github, scholar, email, edu, home,

Text of the page (random words):
hong xing koven yu hong xing koven yu home publications hong xing koven yu ph d in computer science stanford university email koven cs stanford edu google scholar github twitter current research topics world models physical scene understanding dynamics simulation i earned my phd at the stanford vision and learning lab in the computer science department of stanford university advised by jiajun wu i spent two wonderful summers at google research working with noah snavely and bill freeman prior to stanford i visited uc san diego to work with manmohan chandraker i study physics grounded world models recreating the physical world and simulating how it responds to physical actions this requires developing next generation ai systems beyond simply scaling up generative models which lacks the interactability via physical actions my methodology is integrating physical modeling with generative learning my work focuses on the methodological principles in physics grounded world models and how to apply them to broader domains such as robotics and engineering analysis selected awards jane street graduate fellowship finalist 2025 meshy fellowship 2025 roblox graduate fellowship finalist 2024 siggraph asia best paper award 2023 meta research phd fellowship finalist 2023 nvidia graduate fellowship finalist 2022 2023 qualcomm innovation fellowship 2021 stanford school of engineering soe fellowship 2020 publications show selected show all by date show all by topic current research topics world models physical scene understanding dynamics simulation past research topic person re identification indicates equal contributions