🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

1/3 Interested in improving acc and out distribution robustness? 🐥Simply use mixup + cross-entropy 👇 ⚡️RegMixup: Mixup as a Regularizer Can Surprisingly Improve Accuracy and Out Distribution Robustness Paper/Code: arxiv.org/abs/2206.14502 Oxford Torr Vision Group Five @MetaAI

Check out our work on Catastrophic Overfitting from the feature perspective! We will be presenting at #ICML2022 in the New Frontiers in Adversarial Machine Learning@ICML workshop this Friday! Come talk to me and Amartya Sanyal :)

We will be presenting this work on efficient single-step adversarial training at #ICML2022. Come this Friday to the New Frontiers in Adversarial Machine Learning@ICML workshop to talk to Amartya Sanyal and myself. Looking forward to many interesting discussions! 🤓

I'm super happy our paper got accepted in #NeurIPS2022 You can check it out here - arxiv.org/abs/2202.01181 Code - github.com/pdejorge/N-FGSM Thanks so much to all the amazing collaborators Adel Bibi Riccardo Volpi Amartya Sanyal Philip Torr Gregory Rogez and Puneet Dokania

Oxford Torr Vision Group will be presenting 5 papers in this year's NeurIPS. Congratulations to all members involved!

7 papers, 3 of which are accepted as highlights (top 2.5% of all submissions), in CVPR 23! Congratulations to all authors and collaborators! Jishnu Mukhoti Adel Bibi Ameya P. @ ICML 2025 Puneet Dokania A. Tuan Nguyen Pau de Jorge Kejie Li

Now the code is available too! 🚀Hope this can help improve the way we evaluate segmentation models to make them more reliable 🤖🎯⚖️ Paper: arxiv.org/abs/2303.11298 Code: github.com/naver/relis

🔥 #ECCV2024 Showcase your research on the Analysis and Evaluation of emerging VISUAL abilities and limits of foundation models 🔎🤖👁️ at the EVAL-FoMo workshop 🧠🚀✨ 🔗 sites.google.com/view/eval-fomo… Phillip Isola Saining Xie @chrirupp Oxford Torr Vision Group Berkeley AI Research MIT CSAIL

🔥 Excited to be co-organizing this #ECCV2024 workshop with an outstanding line-up of speakers! 🗣️ 🔎Submit if you got papers with new benchmarks and analyses inspecting Emergent Visual abilities ✔️ or limitations ❌of Foundation Models! 🤖

Our recent work on generative models 🎨🖌️🤖 to evaluate 🧐 and fine-tune 🎯 anomaly segmentation methods has been accepted at #ECCV2024 !! 🚀🥳