In the press: AI can’t see as well as humans, and how to fix it
My work has recently been covered by the press.
ICML'25 Paper: Contour Integration Underlies Human-Like Vision
Our new paper on a large-scale comparison of humans and DNNs has been presented at ICML 2025! We show that many models fail when object contours are fragmented, and that the ability to do contour integration is critical for performance and robustness.
Resources
Link to paper: click me!
Link to X post: click me!
2025
ICML’25 Paper: Contour Integration Underlies Human-Like Vision
Our new paper on a large-scale comparison of humans and DNNs has been presented at ICML 2025! We show that many models fail when object contours are fragment...
In the press: Latent Noise Segmentation
My paper on Latent Noise Segmentation, published at the top conference ICML, has recently been picked up by the Finnish press. I gave an interview to Finnish...
2024
CCN’24 Poster: Current DNNs are Unable to Integrate Visual Information Across Object Discontinuities
I attended Cognitive Computational Neuroscience in Boston this year and presented my new work on how Current DNNs are Unable to Integrate Visual Information ...
ICML’24 Paper: Latent Noise Segmentation!
Our paper on: Latent Noise Segmentation has been accepted at ICML 2024! We show that neural noise can be used for unsupervised segmentation, and doing so rep...
2023
New Pre-Print: Latent Noise Segmentation!
Check out our new pre-print: Latent Noise Segmentation! We show that neural noise can be used for unsupervised segmentation, and doing so reproduces many hal...