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:20260805T205503Z LOCATION:Room 408 A DTSTART;TZID=America/Los_Angeles:20260721T140000 DTEND;TZID=America/Los_Angeles:20260721T153000 UID:siggraph_SIGGRAPH 2026_sess131@linklings.com SUMMARY:Light Transport DESCRIPTION:8DNA: 8D Neural Asset Light Transport by Distribution Learning \n\nWe pre-bake complex light scattering in 3D assets into neural 8D light transport functions by learning distributions from forward path-traced sa mples. Unlike previous far-field methods, our representation remains accur ate under near-field illumination, while delivering lower variance and fas ter infe...\n\n\nLiwen Wu and Haolin Lu (University of California San Dieg o); Bing Xu (University of California San Diego, NVIDIA); Miloš Hašan (NVI DIA); and Ravi Ramamoorthi (University of California San Diego)\n--------- ------------\nA Generalizable Light Transport 3D Embedding for Global Illu mination\n\nWe introduce a scalable 3D embedding that learns global illumi nation directly from 3D scene configurations. This design scales to enviro nments with millions of triangles, enabling the first generalizable GI lea rning on complex, high-fidelity indoor scenes, far beyond prior limits. It bypasses per-sc...\n\n\nBing Xu (University of California San Diego, NVID IA); Mukund Varma T, Cheng Wang, and Tzu-mao Li (University of California San Diego); Lifan Wu and Bartlomiej Wronski (NVIDIA); Ravi Ramamoorthi (Un iversity of California San Diego); and Marco Salvi (NVIDIA)\n------------- --------\nRao-Blackwellized Markov chain Monte Carlo Light Transport\n\nWe introduce a novel Rao-Blackwellization technique for variance reduction i n Markov chain Monte Carlo light transport. We improve both Metropolis Lig ht Transport and the current state-of-the-art, Jump Restore Light Transpor t, consistently outperforming traditional variance reduction, accelerating c...\n\n\nSascha Holl (Max Planck Institute for Informatics, Saarland Uni versity); Gurprit Singh (Advanced Micro Devices (AMD)); and Hans-Peter Sei del (Max Planck Institute for Informatics)\n---------------------\nRagged Neighborhood Attention for Spatiotemporal Neural Denoising of Deep Monte C arlo Renderings\n\nRaNAD is the first spatiotemporal neural denoiser for d eep Monte Carlo renderings. It handles semi-structured deep images with va riable bin counts per pixel through ragged neighborhood attention, and exp loits temporal context across up to seven frames, greatly advancing in bot h quality and temporal ...\n\n\nXianyao Zhang, Gerhard Röthlin, Tunc Ozan Aydin, Farnood Salehi, and Marios Papas (Disney Research Studios)\n------- --------------\nPhotons × Force: Differentiable Radiation Pressure Modelin g\n\nWe present a system for optimizing spacecraft designs under radiation pressure, the force that photons exert on objects. It combines a fast Mon te Carlo simulation, a neural proxy of it, and gradient-based optimization to find spacecraft geometry, or operational parameters that minimize trav el time o...\n\n\nCharles Constant and Santosh Bhattarai (University Colle ge London (UCL)), Elizabeth Bates (Alan Turing Institute), and Marek Zieba rt and Tobias Ritschel (University College London (UCL))\n---------------- -----\nDifferentiable Neutron Transport\n\nWe present a differentiable Mon te Carlo framework for multi-energy group neutron transport, extending ide as from differentiable rendering to nuclear engineering. The method enable s gradient-based sensitivity analysis and inverse design.\n\n\nXi Deng, Ma osen Tang, Michael Czekanski, David Bindel, and Steve Marschner (Cornell U niversity)\n---------------------\nLight Transport - Interactive Discussio n\n\nAfter the summary presentations, attendees will participate in an int eractive discussion. Outside the room will be a series of poster boards fo r authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversa tion such...\n\n\nInterest Area: Research & Education\n\nKeyword: Artifici al Intelligence/Machine Learning, Modeling, Rendering, Simulation\n\nRegis tration Category: Full Conference Supporter, Full Conference\n\nSession Ch air: Kun Xu (Tsinghua University) END:VEVENT END:VCALENDAR