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:20260417T190052Z
LOCATION:West Building\, Rooms 211-214
DTSTART;TZID=America/Los_Angeles:20250811T090000
DTEND;TZID=America/Los_Angeles:20250811T103000
UID:siggraph_SIGGRAPH 2025_sess102@linklings.com
SUMMARY:Linear & Non-Linear AI Designs
DESCRIPTION:EditDuet: A Multi-Agent System for Video Non-Linear Editing\n\
 nWe automate video nonlinear editing using a multi-agent system. An Editor
  agent uses tools to create sequences from clips and instructions, while a
  Critic agent provides feedback in natural language. Our learning-based ap
 proach enhances agent communication. Evaluations with an LLM-as-a-judge me
 tric ...\n\n\nMarcelo Sandoval-Castañeda (TTIC, Adobe Research); Bryan Rus
 sell and Josef Sivic (Adobe Research); Gregory Shakhnarovich (TTIC); and F
 abian David Caba Heilbron (Adobe Research)\n---------------------\nOrder M
 atters: Learning Element Ordering for Graphic Design Generation\n\nWe prop
 ose a Generative Order Learner (GOL) that optimizes element ordering for g
 raphic design generation. Our approach learns a content-aware neural order
 , which can significantly improve graphic generation quality, generalize a
 cross different types of generative models and help design generators s...
 \n\n\nBo Yang and Ying Cao (ShanghaiTech University)\n--------------------
 -\nAssetDropper: Asset Extraction via Diffusion Models with Reward-Driven 
 Optimization\n\nAssetDropper is a novel framework for extracting standardi
 zed assets from reference images, addressing challenges such as occlusion 
 and distortion. Leveraging both synthetic and real-world datasets, along w
 ith a reward-driven feedback mechanism, it achieves state-of-the-art perfo
 rmance in asset extr...\n\n\nLanjiong Li (The Hong Kong University of Scie
 nce and Technology (Guangzhou)); Guanhua Zhao (School of Electronic and Co
 mputer Engineering, Peking University); Lingting Zhu (The University of Ho
 ng Kong); Zeyu Cai (The Hong Kong University of Science and Technology (Gu
 angzhou)); Lequan Yu (The University of Hong Kong); Jian Zhang (School of 
 Electronic and Computer Engineering, Peking University); and Zeyu Wang (Th
 e Hong Kong University of Science and Technology (Guangzhou), The Hong Kon
 g University of Science and Technology)\n---------------------\nDesignMana
 ger: An Agent-Powered Copilot for Designers to Integrate AI Design Tools i
 nto Creative Workflows\n\nDesignManager is an AI-powered design support sy
 stem that functions as an interactive copilot throughout the creative work
 flow. With node-based visualization of design evolution and conversational
  interaction modes, it helps designers track, modify, and branch their pro
 cesses while providing context...\n\n\nWeitao You, Yinyu Lu, Zirui Ma, Nan
  Li, Mingxu Zhou, Xue Zhao, Pei Chen, and Lingyun Sun (Zhejiang University
 )\n---------------------\nHybrid Tours: A Clip-based System for Authoring 
 Long-take Touring Shots\n\nWe propose Hybrid Tours, a hybrid approach to c
 reating long-take shots by combining short video clips in a virtual interf
 ace. We show that Hybrid Tours makes capturing long-take touring shots muc
 h easier, and that clip-based authoring and reconstruction lead to higher-
 fidelity results at lower compu...\n\n\nXinrui Liu, Longxiulin Deng, and A
 be Davis (Cornell University)\n---------------------\nLinear & Non-Linear 
 AI Designs - Interactive Discussion\n\nAfter the summary presentations, at
 tendees will participate in an interactive discussion. Outside the room wi
 ll be a series of poster boards for authors to gather around with the audi
 ence. Authors are invited to bring any material related to their paper tha
 t could instigate further conversation such...\n\n---------------------\nM
 oVer: Motion Verification for Motion Graphics Animations\n\nLarge vision-l
 anguage models often fail to capture spatio-temporal details in text-to-an
 imation tasks. We introduce MoVer, a verification system using first-order
  logic to check properties like timing and positioning in motion graphics 
 animations. Integrated into an LLM pipeline, MoVer enables itera...\n\n\nJ
 iaju Ma and Maneesh Agrawala (Stanford University)\n\nInterest Area: Resea
 rch & Education\n\nRecording: Livestreamed, Not Livestreamed, Recorded, No
 t Recorded\n\nRegistration Category: Full Conference, Virtual Access, Mond
 ay\n\nSession Chair: Yotam Gingold (George Mason University)
END:VEVENT
END:VCALENDAR