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:20260417T190057Z
LOCATION:West Building\, Rooms 211-214
DTSTART;TZID=America/Los_Angeles:20250813T090000
DTEND;TZID=America/Los_Angeles:20250813T103000
UID:siggraph_SIGGRAPH 2025_sess128@linklings.com
SUMMARY:Diffusion & Generation
DESCRIPTION:Drag Your Gaussian: Effective Drag-Based Editing  with Score D
 istillation for 3D Gaussian Splatting\n\nDYG is a 3D drag-based scene edit
 ing method for Gaussian Splatting that enables precise, multi-view consist
 ent geometric edits using 3D masks and control points. It combines implici
 t triplane representation and a drag-based diffusion model for high-qualit
 y, fine-grained results. Visit our project pa...\n\n\nYansong Qu, Dian Che
 n, and Xinyang Li (Xiamen University); Xiaofan Li (Baidu Inc.); and Shengc
 huan Zhang, Liujuan Cao, and Rongrong Ji (Xiamen University)\n------------
 ---------\nCMD: Controllable Multiview Diffusion for 3D Editing and Progre
 ssive Generation\n\nCMD revolutionizes 3D generation by enabling flexible 
 local editing of 3D models from a single rendering, as well as progressive
 , interactive creation of complex 3D scenes. At its core, CMD leverages a 
 conditional multiview diffusion model to seamlessly modify/add new compone
 nts—enhancing cont...\n\n\nPeng Li (Hong Kong University of Science and Te
 chnology), Suizhi Ma (Johns Hopkins University), Jialiang Chen and Yuan Li
 u (Hong Kong University of Science and Technology), Congyi Zhang (Univeris
 ty of British Columbia), Wei Xue and Wenhan Luo (Hong Kong University of S
 cience and Technology), Alla Sheffer (Univeristy of British Columbia), Wen
 ping Wang (Texas A&M University), and Yike Guo (Hong Kong University of Sc
 ience and Technology)\n---------------------\nPDT: Point Distribution Tran
 sformation with Diffusion Models\n\nPDT is a novel framework that uses dif
 fusion models to transform unstructured point clouds into semantically mea
 ningful and structured distributions, such as keypoints, joints, and featu
 re lines. Exploring complex point distribution transformation, PDT capture
 s fine-grained geometry and semantics, o...\n\n\nJionghao Wang (Texas A&M 
 University); Cheng Lin (University of Hong Kong); Yuan Liu (HKUST); Rui Xu
  and Zhiyang Dou (University of Hong Kong); Xiaoxiao Long (Nanjing Univers
 ity); Haoxiang Guo (Skywork AI, Kunlun Inc.); Taku Komura (University of H
 ong Kong); and Xin Li and Wenping Wang (Texas A&M University)\n-----------
 ----------\nDiffusion & Generation - Interactive Discussion\n\nAfter the s
 ummary presentations, attendees will participate in an interactive discuss
 ion. Outside the room will be a series of poster boards for authors to gat
 her around with the audience. Authors are invited to bring any material re
 lated to their paper that could instigate further conversation such...\n\n
 ---------------------\nSwiftSketch: A Diffusion Model for Image-to-Vector 
 Sketch Generation\n\nSwiftSketch, a diffusion-based model with a transform
 er-decoder, generates high-quality vector sketches from images in under a 
 second. It progressively denoises stroke coordinates sampled from a Gaussi
 an distribution, effectively generalizing across various object classes. T
 raining uses the ControlS...\n\n\nEllie Arar, Yarden Frenkel, and Daniel C
 ohen-Or (Tel Aviv University); Ariel Shamir (Reichman University); and Yae
 l Vinker (Computer Science and Artificial Intelligence Laboratory (CSAIL),
  Massachusetts Institute of Technology (MIT))\n---------------------\nCLR-
 Wire: Towards Continuous Latent Representations for 3D Curve Wireframe Gen
 eration\n\nCLR-Wire is a unified generative framework for 3D curve-based w
 ireframes, jointly modeling geometry and topology in a continuous latent s
 pace. Using attention-driven VAEs and flow matching, it enables high-quali
 ty, diverse generation from noise, images, or point clouds—advancing CAD d
 esign, sh...\n\n\nXueqi Ma, Yilin Liu, Tianlong Gao, Qirui Huang, and Hui 
 Huang (Shenzhen University)\n---------------------\nRELATE3D: REfocusing L
 atent Adapter for Targeted local Enhancement and Editing in 3D Generation\
 n\nThe alignment of text,images,and 3D is very challenging,yet it is cruci
 al and beneficial for many tasks.We explore and reveal the characteristics
  of the native 3D latent space for 3D generation,make it decomposable and 
 low-rank,thereby enabling efficient learning for multimodal local alignmen
 t,achie...\n\n\nXiao-Lei Li (Tsinghua University, Tencent Video AI Center)
 ; Hao-Xiang Chen (Tsinghua University); Yanni Zhang (Tencent Video AI Cent
 er); Kai Ma (Tencent PCG); Alan Zhao (Tencent Video AI Center); Tai-Jiang 
 Mu (Tsinghua University); Haoxiang Guo (Skywork AI, Kunlun Inc.); and Ran 
 Zhang (Tencent Video AI Center)\n\nInterest Area: Research & Education\n\n
 Recording: Livestreamed, Not Livestreamed, Recorded, Not Recorded\n\nRegis
 tration Category: Full Conference, Virtual Access, Wednesday\n\nSession Ch
 air: Yael Vinker (Massachusetts Institute of Technology (MIT), MIT CSAIL)
END:VEVENT
END:VCALENDAR