BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
X-LIC-LOCATION:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260805T205508Z
LOCATION:Room 403 A
DTSTART;TZID=America/Los_Angeles:20260721T140000
DTEND;TZID=America/Los_Angeles:20260721T153000
UID:siggraph_SIGGRAPH 2026_sess104@linklings.com
SUMMARY:Motion  Generation, Warping & Control
DESCRIPTION:GPC: Large-Scale Generative Pretraining for Transferable Motor
  Control\n\nWe present Generative Pretrained Controllers, a framework for 
 training reusable generative controllers for physically simulated characte
 rs. GPC leverages FSQ discretization with end-to-end RL to learn reusable 
 motor skills from large-scale motion data. Once trained, the generative co
 ntroller can be ...\n\n\nYi Shi (NVIDIA, Simon Fraser University); Yifeng 
 Jiang and Chen Tessler (NVIDIA); and Xue Bin Peng (NVIDIA, Simon Fraser Un
 iversity)\n---------------------\nSMP: Reusable Score-Matching Motion Prio
 rs for Physics-Based Character Control\n\nSMP builds reusable and modular 
 reward models for training motor controllers. Once constructed from a moti
 on dataset, the priors can be reused across diverse tasks while preserving
  the behaviors in the data, without requiring access to the original datas
 et or retraining.\n\n\nYuxuan Mu, Ziyu Zhang, Yi Shi, and Dun Yang (Simon 
 Fraser University); Minami Matsumoto and Kotaro Imamura (Sony Interactive 
 Entertainment); Guy Tevet (Stanford University); Chuan Guo (Snap); Michael
  Taylor (Sony Interactive Entertainment); Chang Shu and Pengcheng Xi (Nati
 onal Research Council Canada); and Xue Bin Peng (Simon Fraser University, 
 NVIDIA)\n---------------------\nMotionBricks: Scalable Real-Time Motions w
 ith Modular Latent Generative Model and Smart Primitives\n\nMotionBricks i
 s a real-time generative framework that transforms interactive motion cont
 rol for animation and robotics. By combining a large-scale latent backbone
  with intuitive "smart primitives," it delivers high-quality, zero-shot mo
 tion synthesis at 15,000 FPS, allowing users to effortlessly bui...\n\n\nT
 ingwu Wang, Olivier Dionne, Michael De Ruyter, David Minor, and Davis Remp
 e (NVIDIA); Kaifeng Zhao (NVIDIA, ETH Zürich); Mathis Petrovich, Ye Yuan, 
 Chenran Li, Zhengyi Luo, Brian Robison, Xavier Blackwell, and Bernardo Ant
 oniazzi (NVIDIA); Xue Bing Peng (NVIDIA, Simon Fraser University); Yuke Zh
 u (NVIDIA, The University of Texas at Austin); and Simon Yuen (NVIDIA)\n--
 -------------------\nDeep Motion Warping via Phase-Conditioned Diffusion A
 utoencoder\n\nWe present a deep motion warping framework utilizing a phase
 -conditioned diffusion autoencoder. By explicitly disentangling motion int
 o root velocity, phase, and style, it enables intuitive, high-level charac
 ter animation editing including root warping, exaggeration, time warping, 
 and style transfer...\n\n\nBowen Zheng (Zhejiang University, Tencent VISVI
 SE); Linjun Wu (Zhejiang University); Xinwei Jiang, Yujin Chai, and Zijiao
  Zeng (Tencent VISVISE); He Wang (University College London (UCL)); and Xi
 aogang Jin (Zhejiang University)\n---------------------\nMultiAct: Text-to
 -Motion Generation from Composite Text via Tailored Attention Guidance\n\n
 Text-to-motion models struggle with composite prompts, often collapsing to
  a single action. We present MultiAct, an inference-time, unpaired framewo
 rk that enhances pretrained generators by amplifying cross-attention for u
 nderrepresented tokens. A lightweight decision scheme selects optimal para
 mete...\n\n\nNathan Sala (Tel Aviv University); Ofir Abramovich and Ariel 
 Shamir (Reichman University); Daniel Cohen-Or (Tel Aviv University); Andre
 as Aristidou (University of Cyprus, CYENS Centre of Excellence); and Sigal
  Raab (Tel Aviv University)\n---------------------\nAutoregressive Diffusi
 on with Hybrid Representation for Interactive Human Motion Generation\n\nA
 RDY is a real-time autoregressive diffusion model for interactive 3D human
  motion generation. Utilizing a novel hybrid motion representation, ARDY e
 mpowers users to interactively synthesize high-fidelity animations directe
 d by dynamic text prompts and flexible spatial constraints, such as full-b
 ody...\n\n\nKaifeng Zhao (ETH Zürich, NVIDIA); Mathis Petrovich, Haotian Z
 hang, and Tingwu Wang (NVIDIA); Siyu Tang (ETH Zürich); and Davis Rempe (N
 VIDIA)\n---------------------\nMotion Generation, Warping & Control - Inte
 ractive Discussion\n\nAfter the summary presentations, attendees will part
 icipate in an interactive discussion. Outside the room will be a series of
  poster boards for authors to gather around with the audience. Authors are
  invited to bring any material related to their paper that could instigate
  further conversation such...\n\n\nInterest Area: Research & Education\n\n
 Keyword: Animation, Artificial Intelligence/Machine Learning, Robotics, Si
 mulation\n\nRegistration Category: Full Conference Supporter, Full Confere
 nce\n\nSession Chair: Yuting Ye (Reality Labs Research, Meta; Meta)
END:VEVENT
END:VCALENDAR