Pulkit (@pulkitkumar95) profile picture

Research Intern @NVIDIA | CS PhD student @umdcs

Joined July 2009
Photo by Pulkit (@pulkitkumar95)Photo by Pulkit (@pulkitkumar95)Photo by Pulkit (@pulkitkumar95)Photo by Pulkit (@pulkitkumar95)Photo by Pulkit (@pulkitkumar95)Photo by Pulkit (@pulkitkumar95)Photo by Pulkit (@pulkitkumar95)Photo by Pulkit (@pulkitkumar95)Photo by Pulkit (@pulkitkumar95)
🧵 VidParse has been accepted to #ECCV2026! 🎉 Can we parse a how-to video as it unfolds - without training? Yes. And it beats the best online supervised baselines, while coming close to offline performance. 🚀 Paper 👇 arxiv.org/abs/2608.27562
2
5
1
31
3,062
Looking forward to presenting Scale Space Diffusion at #CVPR2026! 📍Poster 675 - Poster Session 5, Exhibit Hall F 📅 June 7 (Sunday) 🕒 11:45 AM - 1:45 PM MDT Joint work with @soumikkanad* and @abhi2610. Project webpage: prateksha.github.io/projects…
1
2
1
10
3,386
We’re looking forward to presenting UPLiFT at #CVPR2026! Efficiently extract pixel-dense features from pretrained backbones like DINOv3. We’ll be at the final poster session on Sunday (6/7) from 3:30-5:30pm at Poster 474, so please come by! Website: cs.umd.edu/~mwalmer/uplift/
10
2
75
7,013
Come visit our poster "Scale Space Diffusion" at #CVPR2026! We reformulate diffusion using scale space theory to enable progressive information flow for efficient pixel diffusion! 📍Main Conference: ExHall F 675; Sun, Jun 7, 2026 • 11:45 AM – 1:45 PM MDT
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? 🧵
15
62
5,953
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?
7
30
3
209
19,146
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? 🧵
2
7
1
26
7,292
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/UPLiF… Paper: arxiv.org/abs/2601.17950 Website: cs.umd.edu/~mwalmer/uplift/
2
28
1
148
7,111
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]
8
51
3
388
19,409
🚨 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 [email protected] 📝 Formal UNC CS application required (Deadline soon!) RTs appreciated! 🙏
7
38
1
208
18,410
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!
30
76
13
502
122,082
🧵 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
2
1
2
13
2,038
🌟 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
2
7
3
22
4,820
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.
1
11
4
43
11,988
Call for Papers: #INRV2024 Workshop on Implicit Neural Representation for Vision @ #CVPR2024! Topics: Compression, Representation using INR’s for images, audio, video & more! Ddl: 3/31. Submit now! @CVPR Website: inrv.github.io/ Submission Link: shorturl.at/vzBR8
1
19
1
41
10,886
What are some AI startups/non-profits to follow in India? Either building cool applications or doing some good research. #MachineLearning
1
3
15
3,189
I was watching Lord of the Rings last night and realized there are many parallels with academia! A thread 🧵👇
120
2,511
587
8,240
2,149,971
Excited to share "SparseDet: Improving Sparsely Annotated Object Detection with Pseudo-positive Mining" is accepted to #ICCV2023! We look into training with sparse labels for object detection. w/ @rssaketh, Rama Chellappa and @abhi2610. Website: cs.umd.edu/~sakshams/SparseD…
10
1
53
5,800
(1/N) Our hybrid workshop Open-World Vision will start at 9:00am PDT (6/18). In-person venue: West 118-120 #CVPR23 #cvpr vplow.github.io/vplow_3rd.ht…
1
1
4
423
In case anyone who’s been affected by the recent layoffs finds themselves wanting to work in a more “green” or socially positive space, here’s a great resource 💚 – at Pasadena, CA
🚨New Conservation Tech Directory update!🚨 Now up to 798 resources🎉. Check out the latest here - mailchi.mp/79018b25eb48/janu… @GracieErmi #tech4wildlife
1
1
10
711
At today's @ImageUCPH workshop, section head @KErleben asked me to speak about "how to get cool research ideas." (1/25)
1
57
10
265