CVPR News

📅 #CVPR2026 kicks off June 3 — and so do we. The 12th Medical Computer Vision Workshop opens Day 1 of @CVPR's 43rd edition with 11 keynotes you won't find on any other stage. Room 605 · Colorado Convention Center · Denver @RutgersU · @penn_state · @JohnsHopkins × 2 · @BU_Tweets × 2 · @Stanford · @Cornell · @nvidia · @GoogleResearch · @MSFTResearch Foundation models. Surgical AI. Digital twins. Generative imaging. Interpretable AI. Precision health. Ambient intelligence. Distribution shift. The full map of where medical AI is going — in one room, in one day. @MathiasUnberath @ArchanaVen @hoifungpoon @Jere_je_je @kbatmang @eadeli @mertrory + Dimitris Metaxas · Daguang Xu · Sharon X. Huang · Maddie Traverse #MedicalAI #MedicalComputerVision #ComputerVision #HealthcareAI #CVPR2026

Classic computer animation techniques from the 90s, as used in Toy Story. v/@0xmitsurii

Changing it up this year. Attending #ICRA2026 in Vienna. First time ever attending ICRA!

Attending #CVPR2026 Denver? Before attending, take a moment to read our code of conduct policy.

Did you know that Gaussian splatting always assumed that none of the Gaussians overlap, which is definitely not a correct assumption? Our CVPR 2026 (findings) effort Softmax-GS improves over the non-overlapping assumption in Gaussian splatting, and make multiple Gaussians compete in regions where they overlap using a softmax function, which greatly improves rendering performance when less Gaussians are used, as well as rendering consistency from different viewpoints. Especially, check out our experiments fitting multiple geometric shapes with only 4 Gaussians! arthurhero.github.io/projects/smgs/

I love this figure explaining Bundle Adjustment. It makes all conceptually more clear. @MattNiessner

We are officially wrapping up our calls for ICML. Whether you want to showcase your research or help shape the program, time is running out! Submit your work here: buff.ly/BcmZY8V Apply to be a reviewer here: buff.ly/wULyKtF ⏳ Both deadlines close this Sunday!

🎓 Announcing the 1st Bath Summer School in Machine Learning 📅 5–7 August 2026 | 📍 University of Bath Topics: ODEs/SDEs, Bayesian methods, Self-Supervised Learning, World Models Open to PhDs, postdocs & staff at EU/UK universities bathmlsummerschool.org

Happy to be part of and serve the growth of this amazing community!

BMVC'26 (abs): 2 days + 2h. CORL'26 (abs): 4 days + 14h. CORL'26 (paper): 7 days + 14h. BMVC'26 (paper): 9 days + 2h.

We are so happy to announce our keynote speakers for the Latinx in AI (LXAI) workshop at #CVPR in just 2 weeks! 🌎✨ Please welcome our amazing speakers: Paula Ramos, PhD and Murilo Gustineli, Find more information here: buff.ly/YrMDxlu See you there! 🚀

People are realizing that AIs are nowhere near human intelligence and learning abilities. Yet they have become very useful by compensating for their lack of common sense, lack of understanding of reality, and limited reasoning and planning abilities, by the accumulation of enormous amounts of declarative knowledge.

#CVPR2026 continues to set a global standard for excellence in computer vision and AI research, and this year’s keynotes will dive into new areas of opportunity. 1/2

@JessicaHullman The majority of the papers where we find hallucinations are also single-author, "independent researchers". Many others are from paper mills or sock puppet accounts trying to boost citation counts. You don't want this on arXiv; it is a threat to knowledge.

Announcing the #ICML2026 invited speakers! - Pascale Fung (@pascalefung) - Susan Athey (@Susan_Athey) - Sham Kakade (@ShamKakade6) - Aviv Regev - Verena Rieser (@verena_rieser) - Arvind Narayanan (@random_walker) Check out the blog post for more info!

So proud of my ex-students now outstanding ACs @CVPR #CVPR2026 @mwray0 @bristolcs @BristolUni @davmoltisanti @UniofBath @doughty_hazel @UniLeidenNews This is more important than papers... 3/3 of ex-student ACs are outstanding ACs ☺️

Yann LeCun says LLMs are intrinsically unsafe because they cannot be made fully reliable They can still hallucinate or take agentic actions without predicting the consequences properly "coding works because you can verify the output, but real-world tasks are harder to verify"

BMVC'26 (abs): 6 days + 2h. CORL'26 (abs): 8 days + 14h. CORL'26 (paper): 11 days. BMVC'26 (paper): 13 days.

EBM are so back! @ylecun has been pointing here for years: AI reasoning needs systems that check structure before they answer. Aleph from @logic_int now leads the major formal reasoning benchmarks – let me explain what it is -> 📺

Happy birthday, Patrick Hanrahan! Hanrahan received the #ACMTuringAward in 2019 for fundamental contributions to 3D computer graphics, and the impact of computer-generated imagery (CGI) in filmmaking and other applications. Read more about Hanrahan: buff.ly/9EcqRrt