Gilberto Ochoa-Ruiz

When ChatGPT Broke an Entire Field: An Oral History | Quanta Magazine Researchers in “natural language processing” tried to tame human language. Then came the transformer.

This is a fantastic oral history of the last 10 years of NLP and AI. www.quantamagazine.org/when-chatgpt...

Have you ever wondered what makes two models different?
We all know the ViT-Large performs better than the Resnet-50, but what visual concepts drive this difference? Our new ICLR 2025 paper addresses this question! nkondapa.github.io/rsvc-page/

[1/10] Is scene understanding solved?

Models today can label pixels and detect objects with high accuracy. But does that mean they truly understand scenes?

Super excited to share our new paper and a new task in computer vision: Visual Jenga!

📄 arxiv.org/abs/2503.21770
🔗 visualjenga.github.io

“Inside arXiv—the Most Transformative Platform in All of Science”

www.wired.com/story/inside...

Vision as LoRA We introduce Vision as LoRA (VoRA), a novel paradigm for transforming an LLM into an MLLM. Unlike prevalent MLLM architectures that rely on external vision modules for vision encoding, VoRA internaliz...

Same here… not going back. So currently just hanging around and hopeful that my students will catch any fancy trends … btw … interesting paper arxiv.org/abs/2503.20680 🙃

Experiential Explanations for Reinforcement Learning Reinforcement Learning (RL) systems can be complex and non-interpretable, making it challenging for non-AI experts to understand or intervene in their decisions. This is due in part to the sequential ...

My lab works hard to do human participant evaluations of our Explainable RL systems. I’m rather proud of experiments we ran in this paper: arxiv.org/abs/2210.04723 (coming soon as a journal publication)

Scalable Benchmarking and Robust Learning for Noise-Free Ego-Motion and 3D Reconstruction from Noisy Video

Xiaohao Xu, @tianyizhang.bsky.social, Shibo Zhao, Xiang Li, Sibo Wang, Yongqi Chen, Ye Li, Bhiksha Raj, Matthew Johnson-Roberson, Sebastian Scherer, Xiaonan Huang

arxiv.org/abs/2501.14319

Below my contact details if you are interested!

We have full living allowance scholarships, the tuition fees are fully covered giving you accces to top-notch faculties in our campus, and we have several opportiunities for double degree diplomas and international sojours! 😊

We have several international projects and possibilities for double degrees. Below some of our research topics!

We are a multi-disciplinary group focused on developing AI tools (computer vision) for computer aided diagnosis and minimally invasive surgery! We also work on novel methods in federated learning, robustness, among others!

We are looking for motivated PhD students to join my group CVINSIDE lab!

Myself and Estefanía Talavera and Binod Bhattarai
will be working to make this a very successful track and we expect your contributions in this important domain! Some of the best papers will be selected for a special issue!

We are super happy as our special track "Trustworthy AI for Computer Assisted Diagnosis and Intervention" has been accepted for the 2025 edition of MIUA or the Conference on Medical Image Understanding and Analysis! 😊

Yiqing Liang, Abhishek Badki, Hang Su, James Tompkin, Orazio Gallo
Zero-Shot Monocular Scene Flow Estimation in the Wild
https://arxiv.org/abs/2501.10357

Working from this nice little corner today 😁

A generalist medical language model for disease diagnosis assistance - Nature Medicine Trained on a large corpus of medical text and patient records and tested across diseases, with specific focus on rare presentations, an open-source medical language model demonstrates higher accuracy ...

A generalist #AI medical LLM model that improves accuracy of physician diagnosis across 8 specialities for common and rare diseases
nature.com/articles/s41...

Teaching my #DeepLearning in #ComputerVision course next semester. Looking for blind spots in my content. Share your favourite deep learning course in the thread.

Course: www.eecs.yorku.ca/~kosta/Cours...

Use of robotics in minimally invasive spine surgery and benefits from robotic assistance - Science online Robotic-assisted spine surgery can perform spine fixation procedures. It can be used to treat a variety of spinal conditions, including degenerative disc disease, spinal stenosis, scoliosis, spondylol...

#roboticsurgery #ai #viral
#fyp #robotics #robots #foryou #airobots
Robotic-assisted spine surgery is a minimally invasive surgical technique that uses a robotic arm to assist the surgeon in performing the delicate tasks involved in spinal fusion surgery.
www.online-sciences.com/robotics/use...

MegaSynth: Scaling Up 3D Scene Reconstruction with Synthesized Data

Hanwen Jiang, Zexiang Xu, @desaixie.bsky.social, Ziwen Chen, @haian-jin.bsky.social, Fujun Luan, Zhixin Shu, Kai Zhang, Sai Bi, Xin Sun, Jiuxiang Gu, Qixing Huang, Georgios Pavlakos, Hao Tan

arxiv.org/abs/2412.14166

🚨 Call for papers still open! 🚨

Special Issue on Innovations in Artificial Intelligence for Medicine and Healthcare

Title: Innovations in Artificial Intelligence for Medicine and Healthcare

Journal Health Information Science and Systems (Springer)
Submit your research!

Your figures when using raster rather than vector graphics 🫣

I was waiting for this one! Uncalibrated SLAM with changing focal length, very nice results :)

I'm doing another AI Art Gallery with @cvprconference.bsky.social in Nashville this year 😍🤖🎨

Submissions for artworks using / about computer vision are open until 9th March 2025

More details: bit.ly/CVPRAIArt25

#CVPR2025 #creativeAI #AIart

#MICCAI2024 participants will have received an invite to submit their feedback on how things went.

We'd love to hear from you on what we did well, and what we could have done even better - this will inform next year's chairs.

Organising #MICCAI2024 came with challenges but also great rewards!

Testing and Evaluation of Health Care Applications of Large Language Models This systematic review characterizes the current performance of large language models in evaluating clinical health care settings, including uniformity, thoroughness, and robustness and proposes a fra...

A very necessary systematic review of the use of large language models in healthcare.

jamanetwork.com/journals/jam...

🚨 Call for papers still open! 🚨

Special Issue on Innovations in Artificial Intelligence for Medicine and Healthcare

Title: Innovations in Artificial Intelligence for Medicine and Healthcare

Journal Health Information Science and Systems (Springer)
Submit your research!

Introducing MegaSaM!

Accurate, fast, & robust structure + camera estimation from casual monocular videos of dynamic scenes!

MegaSaM outputs camera parameters and consistent video depth, scaling to long videos with unconstrained camera paths and complex scene dynamics!

Hey thanks a lot! Amazing place that I have yet to explore!

It seems that I’m going to Seattle in January (my 3rd visit) At least i was able to get some rest during December (I wanted to travel a bit less in 2025 😂)