Meta tags:
author=Feng (Jeff) Liang;

Headings (most frequently used words):

-

Text of the page (most frequently used words):
video (21), 2024 (17), with (13), 2025 (12), 2022 (12), and (11), accepted (11), nbsp (10), 2023 (9), the (9), liang (8), gets (8), checkout (8), cvpr (7), austin (7), feng (7), one (7), paper (7), arxiv (6), supervised (6), diana (6), marculescu (6), models (6), our (6), iclr (5), 2021 (5), code (5), for (5), student (4), data (4), efficient (4), clip (4), supmae (4), mae (4), open (4), demo (4), project (4), page (4), zhang (4), bichen (4), movie (4), weaver (4), work (4), february (4), long (4), may (4), current (3), fortunate (3), university (3), fellowship (3), mlcommons (3), image (3), zhao (3), award (3), time (3), vocabulary (3), peter (3), vajda (3), peizhao (3), kunpeng (3), synthesis (3), flowvid (3), streamv2v (3), multi (3), concept (3), personalization (3), april (3), codes (3), january (3), meta (3), papers (3), get (3), 2026 (3), reviewer (2), iccv (2), eccv (2), icml (2), ece (2), undergraduate (2), have (2), tsinghua (2), engineering (2), qualcomm (2), systems (2), rising (2), stars (2), selected (2), declip (2), supervision (2), text (2), yangguang (2), training (2), from (2), best (2), poster (2), aaai (2), eiw (2), first (2), talk (2), huggingface (2), yinan (2), xiaoliang (2), dai (2), segmentation (2), temporal (2), optical (2), present (2), support (2), translation (2), streaming (2), hou (2), october (2), will (2), intern (2), this (2), summer (2), march (2), june (2), fairy (2), december (2), two (2), moviegen (2), research (2), diffusion (2), immiscible (2), form (2), storytelling (2), generative (2), agents (2), agentic (2), jeff (2), thanks, jon, barron, conferences, neurips, journals, tpami, ijcv, tnnls, service, dennis, menn, upenn, cis, grace, kim, cmu, yang, zhou, worked, these, talented, students, collaborators, mentoring, national, scholarship, ministry, education, china, 2014, 2015, excellent, leader, 2018, innovation, finalist, honors, propose, method, efficiently, train, via, utilizing, widespread, among, presentation, bibtex, junjie, yan, fengwei, jing, shao, wanli, ouyang, yufeng, cui, lichen, exists, everywhere, contrastive, language, pre, paradigm, extends, fully, setting, adding, classification, branch, thereby, enabling, effectively, learn, global, features, golden, labels, masked, autoencoders, are, vision, learners, show, generalist, match, performance, specialist, without, dataset, specific, adaptations, 1hour, chinese, 7min, hang, semantic, mask, adapted, leverage, flow, clue, within, enhance, consistency, guided, 5min, highlight, jia, bin, huang, ishan, misra, licheng, jialiang, wang, taming, imperfect, flows, consistent, real, input, realtime, summit, kurt, keutzer, masayoshi, tomizuka, chenfeng, akio, kodaira, looking, backward, feature, banks, animesh, sinha, samaneh, azadi, felix, juefei, tingbo, zecheng, haoyu, tuning, free, anchored, prompts, publications, granted, july, reality, labs, cvprw, ecv, ijcai, oral, three, workshops, august, ovseg, november, gen, again, instruction, based, after, being, rejected, four, times, finally, edge, intelligence, workshop, honored, been, chosen, super, excited, team, push, boundary, personalized, generation, extending, started, full, scientist, speeds, across, tasks, improved, developing, pai, pro, source, agent, framework, phygdpo, news, twitter, zhihu, linkedin, google, scholar, email, their, applications, multimodal, horizon, building, system, utopai, studios, obtained, phd, prior, that, completed, master, bachelor, interests, include, huazhong, science, technology, prof,

Text of the page (random words):
jeff liang feng jeff liang 梁丰 i am building agentic ai video system at utopai studios i obtained my phd from ut austin with prof diana marculescu prior to that i completed my master s at tsinghua university and my bachelor s at huazhong university of science and technology my research interests include agentic ai multimodal agents long horizon agents long form storytelling generative ai image video generative models and their applications email nbsp nbsp cv nbsp nbsp google scholar nbsp nbsp linkedin nbsp nbsp zhihu nbsp nbsp twitter news june 2026 two papers immiscible and phygdpo get accepted to eccv 2026 may 2026 developing pai pro the first open source video agent framework for long form storytelling may 2025 checkout our improved immiscible diffusion speeds up diffusion training 4 across models and tasks april 2025 i started my full time as a research scientist at meta february 2025 one paper movie weaver gets accepted to cvpr 2025 january 2025 one paper streamv2v gets accepted to iclr 2025 december 2024 checkout movie weaver extending moviegen to multi concept personalization october 2024 checkout moviegen super excited to work with the team to push the boundary of personalized video generation may 2024 checkout our streamv2v with code demo may 2024 honored to have been chosen as 2024 mlcommons ml and systems rising stars february 2024 two papers flowvid and fairy get accepted to cvpr 2024 january 2024 after being rejected four times supervised mae supmae finally gets accepted in aaai edge intelligence workshop eiw 2024 with best poster award december 2023 checkout our video to video synthesis work flowvid and instruction based fairy march 2023 i will intern at meta gen ai this summer fortunate to work with dr bichen wu again february 2023 one paper gets accepted to cvpr 2023 november 2022 checkout our open vocabulary segmentation ovseg with codes and demo august 2022 checkout our supervised mae supmae with codes models june 2022 three papers get accepted to icml workshops 2022 april 2022 one paper gets accepted to ijcai 2022 as long oral march 2022 one paper gets accepted to cvprw ecv 2022 february 2022 i will intern at meta reality labs this summer fortunate to work with dr bichen wu january 2022 one paper gets accepted to iclr 2022 october 2021 checkout our data efficient clip declip with codes models july 2021 one paper gets accepted to iccv 2021 april 2021 i am granted ut austin engineering fellowship selected publications movie weaver tuning free multi concept video personalization with anchored prompts feng liang haoyu ma zecheng he tingbo hou ji hou kunpeng li xiaoliang dai felix juefei xu samaneh azadi animesh sinha peizhao zhang peter vajda diana marculescu cvpr 2025 project page arxiv we present movie weaver to support multi concept video personalization looking backward streaming video to video translation with feature banks feng liang akio kodaira chenfeng xu masayoshi tomizuka kurt keutzer diana marculescu iclr 2025 project page arxiv code huggingface demo talk at realtime video ai summit 2025 we present streamv2v to support real time video to video translation for streaming input flowvid taming imperfect optical flows for consistent video to video synthesis feng liang bichen wu jialiang wang licheng yu kunpeng li yinan zhao ishan misra jia bin huang peizhao zhang peter vajda diana marculescu cvpr 2024 highlight project page arxiv 5min video we leverage the temporal optical flow clue within video to enhance the temporal consistency for text guided video to video synthesis open vocabulary semantic segmentation with mask adapted clip feng liang bichen wu xiaoliang dai kunpeng li yinan zhao hang zhang peizhao zhang peter vajda diana marculescu cvpr 2023 project page arxiv code huggingface demo 7min video 1hour talk chinese for the first time we show open vocabulary generalist models match the performance of supervised specialist models without dataset specific adaptations supmae supervised masked autoencoders are efficient vision learners feng liang yangguang li diana marculescu aaai eiw 2024 best poster award arxiv code award supmae extends mae to a fully supervised setting by adding a supervised classification branch thereby enabling mae to effectively learn global features from golden labels supervision exists everywhere a data efficient contrastive language image pre training paradigm yangguang li feng liang lichen zhao yufeng cui wanli ouyang jing shao fengwei yu junjie yan iclr 2022 arxiv bibtex code video presentation we propose data efficient clip declip a method to efficiently train clip via utilizing the widespread supervision among the image text data selected honors mlcommons ml and systems rising stars by mlcommons 2024 qualcomm innovation fellowship finalist by qualcomm 2024 ut austin engineering fellowship by ut austin 2021 2023 excellent student leader by tsinghua university 2018 national scholarship by ministry of education of china 2014 2015 mentoring i m fortunate to have worked with these talented students and collaborators yang zhou undergraduate ut austin ph d student cmu ece current grace kim undergraduate ut austin ph d student upenn cis current dennis menn ph d student ut austin ece current service reviewer of journals tpami ijcv tnnls reviewer of conferences cvpr 2023 2024 2025 iccv 2023 neurips 2023 2024 iclr 2024 2025 eccv 2024 icml 2025 thanks to jon barron