Meta tags:
description=Wenli Xiao is a Robotics PhD student at Carnegie Mellon University, working on humanoid robots, robot foundation models, and agile mobility.;
keywords=robotics,humanoid,CMU,robot learning,reinforcement learning,foundation models,Wenli Xiao;

Headings (most frequently used words):

research, intern, authors, self, improving, vision, language, action, models, with, data, generation, via, residual, rl, hover, versatile, neural, whole, body, controller, for, humanoid, robots, anycar, to, anywhere, learning, universal, dynamics, model, wenli, xiao, news, projects, background,

Text of the page (most frequently used words):
and (16), wenli (11), xiao (11), #research (9), model (9), humanoid (9), for (8), prof (7), twitter (7), guanya (7), shi (7), yuke (6), zhu (6), jim (6), fan (6), website (6), models (6), lead (6), hover (6), luo (5), intern (5), 2025 (5), the (5), vision (5), advisor (4), 2022 (4), lab (4), robotics (4), code (4), arxiv (4), anycar (4), generalist (4), dynamics (4), mobility (4), can (4), via (4), icra (4), haoru (4), xue (4), neural (4), body (4), robot (4), robots (4), linxi (4), zhengyi (4), lin (4), tairan (4), language (4), action (4), cmu (4), 2026 (3), all (3), john (3), dolan (3), 2024 (3), authors (3), manipulation (3), post (3), training (3), with (3), foundation (3), phd (3), aug (2), march (2), scholar (2), nvidia (2), gear (2), background (2), 136 (2), video (2), vehicle (2), agile (2), adapt (2), various (2), cars (2), tasks (2), envs (2), context (2), adaptation (2), outperforming (2), well (2), tuned (2), corl (2), embodiment (2), workshop (2), anywhere (2), learning (2), universal (2), dvij (2), kalaria (2), tony (2), tao (2), 714 (2), parameter (2), network (2), control (2), takes (2), lot (2), subconscious (2), processing (2), humans (2), walk (2), maintain (2), balance (2), maneuver (2), our (2), arms (2), legs (2), into (2), desired (2), positions (2), capture (2), this (2), subconsciousness (2), single (2), that (2), learns (2), how (2), coordinate (2), motors (2), support (2), locomotion (2), versatile (2), whole (2), controller (2), xiaolong (2), wang (2), changliu (2), liu (2), jan (2), kautz (2), zhenyu (2), jiang (2), zhengjia (2), toru (2), pld (2), probe (2), learn (2), distill (2), plug (2), play (2), recipe (2), vla (2), agnostic (2), supporting (2), both (2), autoregressive (2), diffusion (2), architectures (2), push (2), success (2), rates (2), iclr (2), self (2), improving (2), data (2), generation (2), residual (2), jimmy (2), fengyuan (2), yuqi (2), xie (2), andy (2), peng (2), haotian (2), news (2), student (2), blog (2), rights, reserved, bing, jianwei, huang, 2020, ncel, shenzhen, airs, tianjun, zhang, joseph, gonzalez, may, rise, berkeley, yiwei, lyu, june, 2023, carnegie, mellon, univ, institute, summer, riss, present, service, education, experience, projects, received, presidential, fellowship, oct, github, google, email, scientist, doing, advised, build, scalable, general, intelligent, real, world, generally, interested, dexterous, school, computer, science, swipe, browse,

Text of the page (random words):
wenli xiao cmu robotics phd wenli xiao blog research foundation models humanoid mobility all co lead anycar to anywhere learning universal dynamics model corl 2024 x embodiment workshop icra 2025 anycar is a generalist vehicle dynamics model for agile mobility it can adapt to various cars tasks and envs via in context adaptation outperforming well tuned generalist models up to 54 wenli xiao haoru xue tony tao dvij kalaria john dolan guanya shi arxiv website code 136 video twitter co lead hover versatile neural whole body controller for humanoid robots icra 2025 hover is a 1 5m parameter neural network to control the body of a humanoid robot it takes a lot of subconscious processing for us humans to walk maintain balance and maneuver our arms and legs into desired positions we capture this subconsciousness in hover a single model that learns how to coordinate the motors of a humanoid robot to support locomotion and manipulation tairan he wenli xiao toru lin zhengyi luo zhengjia xu zhenyu jiang jan kautz changliu liu guanya shi xiaolong wang linxi jim fan yuke zhu arxiv website code 714 twitter co lead 1 3 self improving vision language action models with data generation via residual rl iclr 2026 pld probe learn distill is a plug and play recipe for vision language action vla post training it is model agnostic supporting both autoregressive and diffusion architectures and can push success rates to 99 wenli xiao haotian lin andy peng haoru xue tairan he yuqi xie fengyuan hu jimmy wu zhengyi luo linxi jim fan guanya shi yuke zhu website twitter swipe to browse research wenli xiao news research background blog wenli xiao robotics phd student at cmu school of computer science i m a phd student at cmu robotics advised by prof guanya shi my vision is to build scalable general intelligent robots in the real world i am generally interested in dexterous manipulation and humanoid robots i m a research scientist intern at nvidia gear lab doing foundation model post training with dr jim fan and prof yuke zhu email google scholar twitter github news 2025 oct received cmu ri presidential fellowship research projects foundation models humanoid mobility all co lead authors wenli xiao haotian lin andy peng haoru xue tairan he yuqi xie fengyuan hu jimmy wu zhengyi luo linxi jim fan guanya shi yuke zhu self improving vision language action models with data generation via residual rl iclr 2026 tl dr pld probe learn distill is a plug and play recipe for vision language action vla post training it is model agnostic supporting both autoregressive and diffusion architectures and can push success rates to 99 website twitter co lead authors tairan he wenli xiao toru lin zhengyi luo zhengjia xu zhenyu jiang jan kautz changliu liu guanya shi xiaolong wang linxi jim fan yuke zhu hover versatile neural whole body controller for humanoid robots icra 2025 tl dr hover is a 1 5m parameter neural network to control the body of a humanoid robot it takes a lot of subconscious processing for us humans to walk maintain balance and maneuver our arms and legs into desired positions we capture this subconsciousness in hover a single model that learns how to coordinate the motors of a humanoid robot to support locomotion and manipulation arxiv website twitter code 714 co lead authors wenli xiao haoru xue tony tao dvij kalaria john dolan guanya shi anycar to anywhere learning universal dynamics model corl 2024 x embodiment workshop icra 2025 tl dr anycar is a generalist vehicle dynamics model for agile mobility it can adapt to various cars tasks and envs via in context adaptation outperforming well tuned generalist models up to 54 arxiv website video twitter code 136 background experience education service research intern nvidia gear lab 2024 present advisor dr jim fan prof yuke zhu research intern carnegie mellon univ robotics institute summer scholar riss june 2022 aug 2023 advisor prof john dolan yiwei lyu research intern rise lab uc berkeley march 2022 may 2022 advisor prof joseph e gonzalez tianjun zhang research intern ncel lab shenzhen airs aug 2020 march 2022 advisor prof jianwei huang prof bing luo 2026 wenli xiao all rights reserved