English
Kosta Derpanis
Report

Kosta Derpanis✓

@CSProfKGD

#CS Associate Prof @YorkUniversity #ComputerVision Researcher, @VectorInst Affiliate @UofT status only, @ELLISforEurope Member, X-social media chair 4all vision

202 Following88.2K Followers41.9K Posts
  1. Kosta Derpanis✓@CSProfKGD
    I’ve launched 2 new course websites for my Introduction to #ComputerVision and #DeepLearning in Computer Vision courses. 🆕 Gaussian Splatting 🆕 Flow Matching 🆕 Sparse Autoencoders 🆕 Lecture Notes (work in progress) The videos don’t include voiceovers, but that’s coming in a future revision 🤞 Share your feedback in the thread.
    Download
    93132745.5K
  2. Kosta Derpanis✓@CSProfKGD
    I vividly recall advice from my PhD supervisor: always read the conference version of a paper before the journal version. Journal versions often become overly formal, losing the intuition of the original paper. This advice has served me well. Today we have a flood of LLM-generated writing with little intuition, excessive formality, and no regard for succinctness. Some papers (and blogs) go on for pages without saying much or, worse, confusing the reader 🙄 Am I against LLMs for writing? NO! But I worry that the next generation of researchers may never develop the skills and judgment to steer them effectively, instead ceding the entire process to an LLM. How can you recognize good scientific writing if you’ve never learned to produce it yourself?
    3120780
  3. Kosta Derpanis✓@CSProfKGD
    Kosta Derpanis
    RT @kwangmoo_yi: Srinivasan et al., "Does VGGT Know What It Does Not Know?" The answer is yes, you can probe the internal features with logistic regression (yes logistic regression) to estimate its failures quite accurately (93.5% AUC).
    04067
  4. Kosta Derpanis✓@CSProfKGD
    Kosta Derpanis
    RT @ARTartaglini: 🟦 Vision transformers emergently learn to segment objects, but how? 100 years ago, Gestalt psychologists had the same question about people. In new work with @Mikey_Lepori, we take inspiration from Gestalt psychology to study how ViTs bind image patches into objects.
    051063
  5. Kosta Derpanis✓@CSProfKGD
    Kosta Derpanis
    RT @holynski_: i’ve found myself using Genie 4 to have conversations with younger versions of myself... here’s me talking to myself just out of college. i was pretty naive, but not a whole lot has changed :)
    072073
  6. Kosta Derpanis✓@CSProfKGD
    Kosta Derpanis
    RT @eigensteve: I really enjoyed taking part in this discussion, especially at this moment in time. Thanks Lex, Alex, and Ankur for the great conversation!
    Lex Fridman @lexfridman
    This is an incredible moment in this history of mathematics, computer science, physics & science with some of the hardest problems in human history being solved by AI over a period of weeks. Depending on your perspective, it's either a crisis, a triumph, or both. I don't think anyone knows for sure. We're all trying to make sense of it. I oscillate between fear/dread and optimism/excitement on a daily basis. The rate of progress is insane, including today's announcement from OpenAI of more problems solved. To help make sense of all of it, here's my conversation with three mathematicians about this moment. Each of them is an expert in the field of each of the biggest problems solved: Navier-Stokes, quasi-Riemann hypothesis, the unique games conjecture and complexity theory more generally. Ankur Moitra is a mathematician and theoretical computer scientist at MIT, visiting the Institute for Advanced Study in Princeton. Steven Brunton (@eigensteve) is a mathematician and fluid dynamics expert at UW. Alex Kontorovich (@AlexKontorovich) is a mathematician at Rutgers specializing in number theory.
    036053
  7. Lex Fridman✓@lexfridman
    This is an incredible moment in this history of mathematics, computer science, physics & science with some of the hardest problems in human history being solved by AI over a period of weeks. Depending on your perspective, it's either a crisis, a triumph, or both. I don't think anyone knows for sure. We're all trying to make sense of it. I oscillate between fear/dread and optimism/excitement on a daily basis. The rate of progress is insane, including today's announcement from OpenAI of more problems solved. To help make sense of all of it, here's my conversation with three mathematicians about this moment. Each of them is an expert in the field of each of the biggest problems solved: Navier-Stokes, quasi-Riemann hypothesis, the unique games conjecture and complexity theory more generally. Ankur Moitra is a mathematician and theoretical computer scientist at MIT, visiting the Institute for Advanced Study in Princeton. Steven Brunton (@eigensteve) is a mathematician and fluid dynamics expert at UW. Alex Kontorovich (@AlexKontorovich) is a mathematician at Rutgers specializing in number theory.
    Download
    2657726K431K
  8. Kosta Derpanis✓@CSProfKGD
    Kosta Derpanis
    RT @janheld14: With 𝐄𝐜𝐡𝐨-𝟑, you can turn the world around you into a 3D space 🌎 I scanned my apartment in less than 5 minutes, and the level of detail in the reconstruction is amazing! 😀 You can explore your space, ask questions about it, redesign it, and much more. Check below! 🤓👇
    SpAItial AI @SpAItial_AI
    Introducing Echo-3. A state-of-the-art world model. Echo-3 turns phone captures into photorealistic, metric-scale 3D worlds. From a single room to multi-floor buildings, outdoor and public spaces. Available today.
    012076
  9. SpAItial AI✓@SpAItial_AI
    Introducing Echo-3. A state-of-the-art world model. Echo-3 turns phone captures into photorealistic, metric-scale 3D worlds. From a single room to multi-floor buildings, outdoor and public spaces. Available today.
    Download
    972381.9K294K
  11. Kosta Derpanis✓@CSProfKGD
    Kosta Derpanis
    RT @KostasPenn: I loved the Informativeness question in the ICLR reviews. This was always the number one question I was asking from reading a paper is : What did I learn? In particular, Nr. 3: "substantially deeper understanding of the problem".
    01083
  15. Kosta Derpanis✓@CSProfKGD
    Kosta Derpanis
    RT @CSProfKGD: Visualization of the day: Slices through visual spacetime
    Download
    legalprimes @LegalPrimes
    Step into the craziest visualization of the fourth dimension you have ever seen
    010015.1K
  16. legalprimes✓@LegalPrimes
    Step into the craziest visualization of the fourth dimension you have ever seen
    Download
    6835.4K47.2K9.8M
  18. Kosta Derpanis✓@CSProfKGD
    Kosta Derpanis
    RT @gulvarol: 📢 PhD position with a flexible start date in 2027. Apply by filling the form if interested:
    039083
  19. Kosta Derpanis✓@CSProfKGD
    Kosta Derpanis
    RT @Rainmaker1973: «Low resolution 3D printed vase» It's an octagon that's been linearly extruded upwards with a bit of rotation every few millimeters. Each "layer" of the vase is scaled (width-wise) based on a sine wave. [📹 risakable]
    0144078
  20. Kosta Derpanis✓@CSProfKGD
    Kosta Derpanis
    RT @orlitany: if you want to come post-docing with me at the Technion, here's your chance:
    Or Litany @orlitany
    Always wanted to come work with me and my students but lacked funding? You’re out of excuses 🙂 The Azrieli Foundation (@azrielifdn) is sponsoring visiting PhD positions (up to 4 months) and postdocs (up to 3 years).
    07078
  21. Or Litany@orlitany
    Always wanted to come work with me and my students but lacked funding? You’re out of excuses 🙂 The Azrieli Foundation (@azrielifdn) is sponsoring visiting PhD positions (up to 4 months) and postdocs (up to 3 years).
    26426.6K
  22. Kosta Derpanis✓@CSProfKGD
    Kosta Derpanis
    RT @jbhuang0604: How thousands of GPUs train one AI Data parallelism helps, but not sufficient. We need new ways to split the workload along different axes. Let's learn about 5D (data, tensor, context, pipeline, experts) parallelism and how they work together.
    053074
  23. Kosta Derpanis✓@CSProfKGD
    "Mechanistic Interpretability” Trailer for my Mech Interp lecture, coming this winter to my #DeepLearning course 🎬
    Download
    54383.2K
  24. Kosta Derpanis✓@CSProfKGD
    Posts @CSProfKGD
    Course:
    001545