Pulkit

Excited to announce my first preprint in LM interpretability! Latent reasoning models are not monitorable by default, since they don't reason in human-readable, natural language text. But can we make progress in understanding their intermediate reasoning steps using mech interp?

Diffusion models be like: “this image is 97% noise… better process all 256×256 pixels anyway” If very noisy diffusion states contain no more useful information than a tiny downsampled image, Then why run expensive full-res computation on them? 🧵

Excited to announce that UPLiFT has been accepted to #CVPR2026! You can also try out UPLiFT right now to extract pixel-dense DINOv3 features with our pretrained models linked below! Code: github.com/mwalmer-umd/UP… Paper: arxiv.org/abs/2601.17950 Website: cs.umd.edu/~mwalmer/uplif…

We’re excited to announce UPLiFT, our lightweight, pixel-dense feature upsampler. UPLiFT boosts feature density, preserves semantics, and has better efficiency scaling than recent SOTA methods. See all links in the thread below. Coauthors: @_sakshams_ @AnirudAgg @abhi2610 🧵[1/6]

🚨 PhD Opening: 3D/4D SLAM & Inverse Rendering for Endoscopy (NIH funded) at @UNC_CS! 🚨 The Project: 🟩 Ideal for students with a strong background in 3D Computer Vision & interested in medical robotics & visualization. 🟩 Highly collaborative work with roboticists, medical imaging researchers, and clinicians (GI, ENT, Pulmonology). 🟩Aim to publish in top-tier CV/ML/Robotics/Medical Imaging venues. 🟩 Build on our prior works: 1⃣Endoscopy depth estimator: PPSNet [ECCV'24 - ppsnet.github.io] 2⃣ 3D SLAM for endoscopy: NFL-BA [NeurIPS'25 -asdunnbe.github.io/NFL-BA/] Apply now! 📧 Email CV to ronisen@cs.unc.edu 📝 Formal UNC CS application required (Deadline soon!) RTs appreciated! 🙏

Aloha #ICCV2025! 🌺 Join @ShuaiyiH, @abhi2610 and me today (21 Oct) at poster #334, 11 AM–1 PM! We will be presenting Trokens — our work on point tracking for action recognition. Details of the paper are below 👇

Everyone says they want general-purpose robots. We actually mean it — and we’ll make it weird, creative, and fun along the way 😎 Recruiting PhD students to work on Computer Vision and Robotics @umdcs for Fall 2026 in the beautiful city of Washington DC!

Looking forward to presenting this work and discussing it with fellow researchers. See you in Hawaii! 🌺🌺 #ICCV25 #ICCV2025 #ICCV

👥 This work was co-led with @ShuaiyiH in collaboration with @MatthewWalmer, @rssaketh , and @abhi2610 🔗 All code and data are released: 🌐 Webpage: trokens-iccv25.github.io 📄 ArXiv: arxiv.org/abs/2508.03695 💻 Code: github.com/pulkitkumar95/… 🤗 Data: huggingface.co/datasets/pulki…

🎉 Excited to share our paper "Trokens: Semantic-Aware Relational Trajectory Tokens for Few-Shot Action Recognition" has been accepted to #ICCV2025! Equally co-led with @ShuaiyiH — we advance few-shot action recognition via smart point tracking. 🔗 trokens-iccv25.github.io 🧵👇

@zenwill_ai @ICCVConference From the instructions

@ICCVConference In "Camera-Ready Submission Instructions", it says "However, papers that are longer than 8 pages (not including references), will not be processed and will not appear in the conference proceedings or on IEEE Xplore." Does this "8 pages" also not include acknowledgment section？

If copyright status and paper status are “received” or “submitted” then ignore email.

🧵 Your DiT, faster Introducing ECAD: we reframe diffusion model caching as multi-objective optimization and evolve Pareto-optimal schedules via a genetic algorithm—achieving 4.47 FID gain at 2.58× speedup, with no retraining or tuning. 🔗 aniaggarwal.github.io/ecad #MachineLearning

🌟 CoLLM: A Large Language Model for Composed Image Retrieval (CVPR 2025) ✨A cutting-edge training paradigm using image-caption pairs 📊High-quality synthetic triplets for training & benchmarking 🔗Project: collm-cvpr25.github.io 📄Paper: arxiv.org/abs/2503.19910 #LLM #CIR

How can we make Imitation Leaning generalize? In my latest work we show that a key point based representation can generalize to novel instances of an object and is agnostic to background changes.

Finally, I found some time over the break to clean the code up and release it: github.com/pulkitkumar95/… :)