Isabella Yu
@yukaryote.bsky.social
PhD @Berkeley prev CS/Math @MIT
Reposted by Isabella Yu
Happy to finally share our latest work on Dataset Distillation!
"Dataset Distillation for Pre-Trained Self-Supervised Vision Models," set to appear at #NeurIPS 2025!
We learn 1 image per class to train linear heads for pre-trained models.
linear-gradient-matching.github.io
More in thread 🔽
"Dataset Distillation for Pre-Trained Self-Supervised Vision Models," set to appear at #NeurIPS 2025!
We learn 1 image per class to train linear heads for pre-trained models.
linear-gradient-matching.github.io
More in thread 🔽
Dataset Distillation for Pre-Trained Self-Supervised Vision Models
Distilling image datasets to a single sample per class to train classifiers on top of pre-trained self-supervised vision models.
linear-gradient-matching.github.io
November 21, 2025 at 4:01 AM
Happy to finally share our latest work on Dataset Distillation!
"Dataset Distillation for Pre-Trained Self-Supervised Vision Models," set to appear at #NeurIPS 2025!
We learn 1 image per class to train linear heads for pre-trained models.
linear-gradient-matching.github.io
More in thread 🔽
"Dataset Distillation for Pre-Trained Self-Supervised Vision Models," set to appear at #NeurIPS 2025!
We learn 1 image per class to train linear heads for pre-trained models.
linear-gradient-matching.github.io
More in thread 🔽
Fan art (?) for The Glass Castle by Jeannette Walls, anyone else read this in high school #procreate
November 16, 2025 at 10:41 AM
Fan art (?) for The Glass Castle by Jeannette Walls, anyone else read this in high school #procreate
Reposted by Isabella Yu
DUSt3R et al. are impressive, but how do they actually work? We investigate this in our project 𝘜𝘯𝘥𝘦𝘳𝘴𝘵𝘢𝘯𝘥𝘪𝘯𝘨 𝘔𝘶𝘭𝘵𝘪-𝘝𝘪𝘦𝘸 𝘛𝘳𝘢𝘯𝘴𝘧𝘰𝘳𝘮𝘦𝘳𝘴!
We share findings on the iterative nature of reconstruction, the roles of cross and self-attention, and the emergence of correspondences across the network [1/8] ⬇️
We share findings on the iterative nature of reconstruction, the roles of cross and self-attention, and the emergence of correspondences across the network [1/8] ⬇️
𝗨𝗻𝗱𝗲𝗿𝘀𝘁𝗮𝗻𝗱𝗶𝗻𝗴 𝗠𝘂𝗹𝘁𝗶-𝗩𝗶𝗲𝘄 𝗧𝗿𝗮𝗻𝘀𝗳𝗼𝗿𝗺𝗲𝗿𝘀
Michal Stary, Julien Gaubil, Ayush Tewari, Vincent Sitzmann
arxiv.org/abs/2510.24907
Trending on www.scholar-inbox.com
Michal Stary, Julien Gaubil, Ayush Tewari, Vincent Sitzmann
arxiv.org/abs/2510.24907
Trending on www.scholar-inbox.com
November 4, 2025 at 7:40 PM
DUSt3R et al. are impressive, but how do they actually work? We investigate this in our project 𝘜𝘯𝘥𝘦𝘳𝘴𝘵𝘢𝘯𝘥𝘪𝘯𝘨 𝘔𝘶𝘭𝘵𝘪-𝘝𝘪𝘦𝘸 𝘛𝘳𝘢𝘯𝘴𝘧𝘰𝘳𝘮𝘦𝘳𝘴!
We share findings on the iterative nature of reconstruction, the roles of cross and self-attention, and the emergence of correspondences across the network [1/8] ⬇️
We share findings on the iterative nature of reconstruction, the roles of cross and self-attention, and the emergence of correspondences across the network [1/8] ⬇️
Machines of loving grace #procreate #illustration
October 20, 2025 at 9:59 AM
Machines of loving grace #procreate #illustration
Reposted by Isabella Yu
Wow!
So while I still wish this case would’ve gone to trial, and believe the amount given per work should have been EVEN higher, $1.5 billion in settlement AND the destruction of infringing datasets is still a fantastic start! Congratulations to all those involved as these things are not easy!
So while I still wish this case would’ve gone to trial, and believe the amount given per work should have been EVEN higher, $1.5 billion in settlement AND the destruction of infringing datasets is still a fantastic start! Congratulations to all those involved as these things are not easy!
BREAKING: Anthropic has agreed to pay at least $1.5 billion to authors in class action lawsuit for pirating their works www.wired.com/story/anthro...
Anthropic Agrees to Pay Authors at Least $1.5 Billion in AI Copyright Settlement
Anthropic will pay at least $3,000 for each copyrighted work that it pirated. The company downloaded unauthorized copies of books in early efforts to gather training data for its AI tools.
www.wired.com
September 5, 2025 at 9:00 PM
Wow!
So while I still wish this case would’ve gone to trial, and believe the amount given per work should have been EVEN higher, $1.5 billion in settlement AND the destruction of infringing datasets is still a fantastic start! Congratulations to all those involved as these things are not easy!
So while I still wish this case would’ve gone to trial, and believe the amount given per work should have been EVEN higher, $1.5 billion in settlement AND the destruction of infringing datasets is still a fantastic start! Congratulations to all those involved as these things are not easy!
Reposted by Isabella Yu
Really excited to share this with you all! The website is already live at anadodik.github.io/publication/...
August 5, 2025 at 3:59 PM
Really excited to share this with you all! The website is already live at anadodik.github.io/publication/...
Reposted by Isabella Yu
MIT News today features a fun SIGGRAPH paper by GDP student Ana (@geometry.gay) collaboarting w/ students @yukaryote.bsky.social @kartikchandra.bsky.social plus us old folks. This work explores what it means to do geometry processing on Escher-like "impossible" shapes! news.mit.edu/2025/mit-mes...
MIT tool visualizes and edits “physically impossible” objects
The “Meschers” tool from MIT CSAIL represents “physically impossible” objects commonly found in M.C. Escher’s illustrations by converting both images and 3D models in 2.5-dimensional objects. The tool...
news.mit.edu
August 5, 2025 at 1:07 PM
MIT News today features a fun SIGGRAPH paper by GDP student Ana (@geometry.gay) collaboarting w/ students @yukaryote.bsky.social @kartikchandra.bsky.social plus us old folks. This work explores what it means to do geometry processing on Escher-like "impossible" shapes! news.mit.edu/2025/mit-mes...
April 11, 2025 at 4:09 AM