Alejandro Pardo

🚀 Our MatchDiffusion was accepted to ICCV 2025 in Hawaii! 🌺 We generate two synchronized videos from text prompts—designed for match-cuts. Results: matchdiffusion.github.io Paper: arxiv.org/abs/2411.18677 #MatchDiffusion #ICCV2025 #DiffusionModels #TextToVideo #GenerativeAI

LiteLLM HAS BEEN COMPROMISED, DO NOT UPDATE. We just discovered that LiteLLM pypi release 1.82.8. It has been compromised, it contains litellm_init.pth with base64 encoded instructions to send all the credentials it can find to remote server + self-replicate. link below

Consejos si estás en Venezuela: 1. Mantén un perfil bajo y prioriza tu seguridad. 2. Evita zonas de protestas o presencia armada. 3. Muévete solo si es imprescindible. Si lo haces, hazlo de día y por vías principales. 4. Acuerda un plan y un punto de contacto con tu familia. 5. Ten el móvil cargado y una copia de tus documentos. 6. Ten agua/medicinas a mano, comida no perecedera y linterna. 7. No difundas audios/cadenas sin confirmar, ni tomes decisiones en base a rumores. 8. Ten tu efectivo en billetes pequeños y en al menos dos lugares distintos. 9 Apunta tus contactos más importantes en papel. 10. Desactiva la vista previa de mensajes con la pantalla bloqueada. 11. Evita publicar en tiempo real dónde estás o por dónde te mueves.

Heading towards Hawaii 🌺✈️ Excited to present MatchDiffusion at #ICCV2025! 🗓️ Wed 22 Oct, 2:45–4:45 p.m. HST 📍 Exhibit Hall I, #1373 📄 Paper: arxiv.org/abs/2411.18677 Poster info: iccv.thecvf.com/virtual/2025/p… #GenerativeAI #MatchCut #ComputerVision See you there!!

A PhD can be incredibly rewarding and fun, under the right circumstances. If you have tremendous *fire* for research and self motivation, it can be the right path for you. If you're doing it because it's the least friction next step after ugrad or MS, reconsider.

🌟Excited to share that our IVUL Lab @KaustVision has seven papers accepted at @ICCVConference 2025! We'll be presenting them this month in Hawaii 🌺🌴 👀 Check out the details in the images below. Hope to see you there! Big thanks to all our amazing collaborators 🤝 #ICCV2025

🎓PhD defense! 🎉Huge congratulations to @MereyRamazanova from the IVUL family on successfully defending her PhD thesis, “Towards Robust Multimodal Egocentric Video Understanding”, advised by Prof. @BernardSGhanem Wishing you all the best in your next chapter, Dr. Merey!🚀

GenAI isn't just a technology; it's an informational pollutant—a pervasive cognitive smog that touches and corrupts every aspect of the Internet. It's not just a productivity tool; it's a kind of digital acid rain, silently eroding the value of all information. Every image is no longer a glimpse of reality, but a potential vector for synthetic deception. Every article is no longer a unique voice, but a soulless permutation of data, a hollow echo in the digital chamber. This isn't just content creation; it's the flattening of the entire vibrant ecosystem of human expression, transforming a rich tapestry of ideas into a uniform, gray slurry of derivative, algorithmically optimized outputs. This isn't just innovation; it's the systematic contamination of our data streams, a semantic sludge that clogs the channels of genuine communication and cheapens the value of human thought—leaving us to sift through a digital landfill for a single original idea.

🎓PhD dissertation season at IVUL! 🎉Congratulations to Alejandro Pardo on successfully defending his #PhD dissertation, “Computer Vision for Video Editing: Learning to Cut, Classify, Assemble, and Generate,” advised by Prof. @BernardSGhanem Great work, Alejo! 🚀👏

On June 23, I defended my Ph.D. in Computer Vision at KAUST! 🎓 My thesis explored AI-driven video editing, and a few days later, MatchDiffusion, one of its core contributions, was accepted to ICCV 2025! Grateful to my advisors & committee. 🚀 #ICCV2025

@MartinNebelong Hi @MartinNebelong, Veo3 never ceases to amaze me. The jump-cut here is awesome, it's impressive the model is able to keep the picture some time on the screen before jump cutting to the other scene. Do you know whether is possible to create match-cuts with it?

Take a look at how consistent my sketch is, despite the change in lighting and the change in the underlying material.. The way the water distort the image. Creating something as realistic as this in traditional 3d would be VERY difficult. Notice how the model stays on the original frame for a short while and then jump/cuts.

Veo has an incredible "hidden" ability. It's one of my absolute favorite aspects of Veo 3 i2v. It's ability to transport elements through the latent space, with this simple but very powerful prompt structure: “Instantly jump/cut on frame 1. [Describe the new context]" With this structure, I was able to use my quick drawing as the main subject in all the scenes you saw here. Seeing how the model is able to retain my personal sketch in all of these different contexts.. Feels magical honestly 😍 Music from artlist.io. My personal opinion about the question asked here, should be clear if you've followed along with my posts 🔥🎨

El Colombiano sale al extranjero y se da cuenta que: - Nuestro sistema de salud era bueno tirando a muy bueno. - Nuestro costo de Internet es baratísimo. - Pese a nuestra violencia, el Colombiano es amable con el desconocido. - Somos privilegiados en gastronomía.

Part 2 of this mystery. Spotted on reddit. In my test not 100% reproducible but still quite reproducible. 🤔

Hello! If you like pretty images and videos and want a rec for CVPR oral session, you should def go to Image/Video Gen, Friday at 9am: I'll be presenting "Motion Prompting" @RyanBurgert will be presenting "Go with the Flow" and @ChangPasca1650 will be presenting "LookingGlass"

That's a wrap for MULA 2025! See you (hopefully) next year! #CVPR2025

🧵 1/8 The Illusion of Thinking: Are reasoning models like o1/o3, DeepSeek-R1, and Claude 3.7 Sonnet really "thinking"? 🤔 Or are they just throwing more compute towards pattern matching? The new Large Reasoning Models (LRMs) show promising gains on math and coding benchmarks, but we found their fundamental limitations are more severe than expected. In our latest work, we compared each “thinking” LRM with its “non-thinking” LLM twin. Unlike most prior works that only measure the final performance, we analyzed their actual reasoning traces—looking inside their long "thoughts". Our analysis reveals several interesting results ⬇️ 📄 machinelearning.apple.com/research/illus… Work led by @ParshinShojaee and @i_mirzadeh, and with @KeivanAlizadeh2, @mchorton1991, Samy Bengio.

#KostasThoughts: As an AC, it’s frustrating when reviewers only see the rebuttal through the lens of their own review. Step back—consider all reviews and discussions. DON’T rush to finalize; see the FULL picture!

LOVE a #CVPR2025 paper but facing pushback from the other reviewers? CHAMPION IT! Make a strong case in your discussions and your “justification of rating”—fight for the papers that deserve it! Don’t be swayed by being in the minority in the face of weak arguments.

The use of AI in reviewing is a growing problem. Several of my ICCV papers have AI reviews -- one reviewer was so lazy that they left in the prompts! A common refrain that I hear is that people have difficulty writing in English and need to use AI to clean up their review. Hogwash. I took one of the reviews I wrote for ICCV and used Google Translate to translate it to German and then from German to Spanish, and back to English. The original English review was rated as 99% human by an AI detector. After the multiple translations, this only dropped to 93% human. So, if English is a problem, write in whatever language you like and use AI to translate it. This should be the only use of AI allowed by the rules. Or, better yet, submit your review in whatever language you feel most comfortable with and have OpenReview automatically translate into whatever language the author wants. So let's get rid of this "grammar argument" for using AI. People are using AI in reviewing to cheat the system. I can submit my own paper to an AI for a review if I want. You've added no value if you do the same. What I want from a reviewer is their unique insight. If they don't understand my paper, that's important information for me. They represent my human audience. If they struggle with my submission, so with other humans. As long as my paper is written for humans, I need feedback from humans.