Meta tags:

Headings (most frequently used words):

news, selected, publications, full, list, invited, talks, and, reports, academic, services, ruihang, chu,

Text of the page (most frequently used words):
wan2 (22), 2025 (18), and (17), video (15), the (13), generation (12), models (11), accepted (10), wan (9), with (9), 2026 (9), #ruihang (8), chu (8), paper (6), long (6), code (6), was (5), reported (5), tech (5), media (5), chen (5), multimodal (5), open (5), neurips (4), iclr (4), 新智元 (4), liu (4), zhang (4), yang (4), new (4), generative (4), architecture (4), prof (4), conference (3), iccv (3), sana (3), performance (3), aware (3), chi (3), cost (3), enable (3), zhao (3), wang (3), han (3), enze (3), xie (3), oral (3), level (3), controllable (3), native (3), foundation (3), papers (3), are (3), launched (3), program (2), reviewer (2), icml (2), aaai (2), pami (2), academic (2), services (2), invited (2), survey (2), mini (2), gemini (2), longlive (2), diverse (2), computer (2), weekly (2), talks (2), reports (2), sota (2), viewpoint (2), metric (2), learning (2), wei (2), training (2), more (2), efficient (2), linear (2), yuyang (2), shuai (2), xianbang (2), yicheng (2), hongwei (2), muyang (2), song (2), diffusion (2), h100 (2), gpu (2), interactive (2), real (2), time (2), completion (2), wing (2), jiaya (2), jia (2), commercial (2), latent (2), trajectory (2), guidance (2), yujiu (2), move (2), via (2), full (2), role (2), multi (2), shot (2), audio (2), visual (2), moe (2), cinematic (2), aesthetics (2), publications (2), sourced (2), main (2), have (2), online (2), source (2), model (2), million (2), features (2), news (2), email (2), world (2), now (2), about (2), research (2), scientist (2), alibaba (2), team (2), supervised (2), homepage (2), committee, cvpr, eccv, iros, journal, games, webinar, talk, algorithm, presentations, official, release, livestreams, 量子位, has, been, many, times, outlets, positively, covered, international, medias, 人民日报海外版, scmp, information, cnbc, reuters, top, accuracy, handle, view, variations, image, retrieval, yifan, sun, yadong, zheng, yichen, vehicle, identification, 2019, faster, latency, competitive, inference, only, moviegen, minute, constant, memory, cache, than, vanilla, attention, dit, junsong, jincheng, junyu, pan, daquan, zhou, huan, ling, haozhe, hao, yukang, cai, sanja, fidler, ping, luo, block, transformer, days, fine, tuning, train, test, 240, second, recache, for, prompts, inserted, fps, single, huang, xiao, yingcong, yao, multiple, partial, shapes, condition, occupancy, fusion, decrease, error, hierarchical, feature, aggregation, shentong, zhenguo, matthias, nießner, diffcomplete, based, shape, 2023, longer, durations, rich, annotations, movebench, benchmark, par, point, control, change, base, i2v, yefei, zhekai, shiwei, xiaogang, bin, xia, dingdong, xihui, hengshuang, yingya, motion, stack, creative, toolkit, sync, structure, hybrid, ti2v, powerful, vae, core, contributor, advanced, large, scale, technical, report, selected, list, reasoning, acm, computing, surveys, acl, inserter, freescale, emnlp, target, dpo, four, generated, videos, downloads, chinese, first, grade, genation, backbone, disco, edit, tops, hugging, face, dec, starring, 15s, length, six, omniverifier, nejm, chexgen, s2l, storytelling, editing, cloning, restyling, continuation, you, interested, collaboration, feel, free, drop, developed, leading, including, welcome, try, call, our, apis, download, vlms, coding, llms, captioning, language, pre, post, work, vision, term, goal, focus, two, key, directions, simulate, reason, alistar, talent, also, hold, joint, postdoctoral, position, tsinghua, university, received, degree, from, cuhk, during, stage, had, pleasure, working, xiaojuan, 储瑞航, google, scholar, github, twitter, hangzhou, china, group,

Text of the page (random words):
ruihang chu homepage homepage about me news publications talks and reports academic services ruihang chu research scientist wan team alibaba group hangzhou china email twitter github google scholar hi i am ruihang chu 储瑞航 i am a research scientist at the alibaba wan team via the alistar talent program i also hold a joint postdoctoral position at tsinghua university supervised by prof yujiu yang i received my ph d degree from cuhk supervised by prof jiaya jia and prof chi wing fu during my ph d stage i had the pleasure of working with prof xiaojuan qi i work on generative ai and computer vision my long term goal is to enable models to simulate and reason about the visual world now i focus on two key directions video generation pre and post training of foundation models native multimodal architecture controllable generation and ar models multimodal language models vlms coding llms and multimodal captioning we have developed world leading video generative foundation models wan now including wan2 1 wan2 2 wan2 5 wan2 6 and wan2 7 welcome to to try it online call our apis or download the open source models if you are interested in collaboration feel free to drop me an email news 2026 04 wan2 7 is launched new features controllable video storytelling editing generation cloning restyling continuation and more 2026 04 s2l po is accepted to icml 2026 2026 03 chexgen is accepted to nejm ai 2026 01 six papers are accepted to iclr 2026 with sana video and omniverifier as oral 2025 12 wan2 6 is launched new features starring role to video multi shot generation 15s generation length 2025 12 wan move tops hugging face weekly papers dec 7 13 2025 12 mini gemini is accepted to t pami 2025 11 o disco edit is accepted to aaai 2026 as oral 2025 09 wan2 5 is launched the chinese first commercial grade video audio genation model with the native multimodal backbone 2025 09 wan models have generated 70 million videos online and 40 million open source model downloads 2025 09 four papers are accepted to neurips 2025 2025 08 target dpo is accepted to emnlp 2025 main conference 2025 07 wan2 2 is open sourced new moe architecture with cinematic level aesthetics 2025 06 freescale is accepted to iccv 2025 2025 05 inserter is accepted to acl 2025 main conference 2025 04 the survey of reasoning with foundation models is accepted to acm computing surveys 2025 02 wan2 1 is open sourced selected publications full list technical report wan open and advanced large scale video generative models paper wan2 1 code wan2 2 code core contributor wan2 1 sota performance powerful video vae wan2 2 moe architecture cinematic level aesthetics hybrid ti2v wan2 5 audio visual sync native multimodal structure wan2 6 role to video multi shot generation wan2 7 full stack creative toolkit neurips 2025 wan move motion controllable video generation via latent trajectory guidance paper code ruihang chu yefei he zhekai chen shiwei zhang xiaogang xu bin xia dingdong wang hongwei yi xihui liu hengshuang zhao yu liu yingya zhang yujiu yang latent trajectory guidance no architecture change on base i2v models point level control on par with commercial models movebench benchmark diverse longer durations rich annotations neurips 2023 diffcomplete diffusion based generative 3d shape completion paper code ruihang chu enze xie shentong mo zhenguo li matthias nießner chi wing fu jiaya jia hierarchical feature aggregation 40 decrease on l1 completion error occupancy aware fusion enable multiple partial shapes as the condition iclr 2026 longlive real time interactive long video generation paper code shuai yang wei huang ruihang chu yicheng xiao yuyang zhao xianbang wang muyang li enze xie yingcong chen yao lu song han real time 20 7 fps generation on a single h100 gpu interactive kv recache for new prompts inserted long train long test long up to 240 second generation fine tuning cost 32 h100 gpu days on wan2 1 1 3b iclr 2026 oral sana video efficient video generation with block linear diffusion transformer paper code junsong chen yuyang zhao jincheng yu ruihang chu junyu chen shuai yang xianbang wang yicheng pan daquan zhou huan ling haozhe liu hongwei yi hao zhang muyang li yukang chen han cai sanja fidler ping luo song han enze xie linear dit more efficient than vanilla attention constant memory kv cache enable minute long video generation training cost only 1 of the cost of moviegen inference 16 faster in latency with competitive performance iccv 2019 vehicle re identification with viewpoint aware metric learning paper ruihang chu yifan sun yadong li zheng liu chi zhang yichen wei viewpoint aware metric learning handle view variations in image retrieval sota performance 10 top 1 accuracy invited talks and reports wan was positively covered by international medias e g reuters cnbc the information computer weekly scmp and 人民日报海外版 wan has been reported many times by diverse tech media outlets longlive was reported by tech media 新智元 sana video was reported by tech media 新智元 mini gemini was reported by tech media 新智元 and 量子位 the survey was reported by tech media 新智元 algorithm presentations in the official release livestreams of wan2 2 wan2 5 wan2 6 and wan2 7 invited talk by games webinar academic services journal reviewer t pami t ip ra l conference program committee reviewer neurips iclr icml cvpr iccv eccv aaai and iros