Researcher at University of Oxford / TU Munich
Munich
Joined February 2009
- Tweets92
- Following350
- Followers393
- Likes279
Pinned Tweet
In-context learning suggests that a model has learned versatile representations. What if we use in-context learning itself as a training task for visual representations?
๐ฃ Introducing ๐๐๐๐: ๐๐ถ๐ป๐ฒ๐ฎ๐ฟ ๐๐ป-๐๐ผ๐ป๐๐ฒ๐
๐ ๐๐ฒ๐ฎ๐ฟ๐ป๐ถ๐ป๐ด โจ @CVPR 2026 Oral โจ
๐๐๐๐ trains on videos without manual annotation.
Key idea: An optimal linear mapping that predicts dense cues (e.g. depth, flow), estimated on one video frame, should also predict the corresponding cues of another frame from the same video.
This yields compelling results on dense vision tasks: video object segmentation, (zero-shot) semantic segmentation and surface normal estimation.
Paper, code, models and demo: lila-pixels.github.io
Joint work with @ma_sundermeyer, Hidenobu Matsuki, David Joseph Tan and @fedassa (and special thanks to David and Federico for hosting my research visit at Google).
#cvpr2026 @Google @MunichCenterML @tumcvg @TU_Muenchen
Nikita Araslanov retweeted
๐ข LucidLyra: Feed-Forward Generative 3D Scene Reconstruction and Joint Understanding via Multi-Task Distillation!
Generate a queryable 3D Gaussian scene from a single image! Our feed-forward decoder jointly predicts geometry, appearance, open-vocabulary semantics, and object instances. We train on generated videos, combining video-model self-distillation with supervision from frozen 2D vision foundation models. LucidLyra enables applications such as physical AI simulation, 3D scene editing, and 3D asset mining.
๐ research.nvidia.com/labs/silโฆ
On the bright side, LLMs have achieved what decades of scientific writing could not: suddenly, papers seem to care about the difference between a hyphen, an en dash, and an em dash.
Nikita Araslanov retweeted
Besides scalability, self-supervised learning is attractive because we have a good real-life example of humans/mammals learning strong visual systems.
BUT: they do not learn from shuffled image collections. They learn from (essentially) a single long visual stream.
This is what we tackle in our new @NeurIPSConf paper, led by Ivan, together with @LukasKnobel1 and me. 1/
arxiv.org/abs/2609.40333
Nikita Araslanov retweeted
DINOv2 has never seen a caption, and Qwen3 has never seen an image. We still aligned their embedding spaces without a single image-caption pair.
It even works when the images and the captions come from different datasets.
Project page: dominik-schnaus.github.io/unโฆ
โฌ๏ธ
Nikita Araslanov retweeted
Attention arXiv authors. Because of the exponential increase in submissions, arXiv is limiting the number of papers that a person can submit each month to two. 1/
arXiv has updated our policy on rate limiting for all submitters.
This update was made to fairly distribute moderator time & support the arXiv community of staff, volunteers, readers & authors.
Please read our announcement to learn more: blog.arxiv.org/2026/10/01/upโฆ
ALT arXiv logo in white over a whiteboard with math symbols.
Nikita Araslanov retweeted
๐จ Apply now for the #ELLISPhD Program!
Work with leading AI/ML researchers across Europe, with exchange visits & joint supervision. The application deadline is 31 Oct.
Get all info & apply now at ellis.eu/news/ellis-phd-progโฆ
Nikita Araslanov retweeted
๐น Job alert: PhD Position in Mathematical Foundations of Deep Learning at TU Darmstadt
๐ Darmstadt ๐ฉ๐ช
โฐ 22 October
๐ bit.ly/4dzSuaG
Nikita Araslanov retweeted
๐ฆ RHINO: Reconstructing Human Interactions with Novel Objects from Monocular Videos
From one moving-camera RGB video, we reconstruct a 3D human, an unseen manipulated 3D object, and the static 3D scene in one world frame
Video input, then 3D output. ๐ Code and data are public!
Nikita Araslanov retweeted
Wrote done my thoughts on what's fundamentally new for vision when using agents --- obviously, apart from some hard problems that have now been nearly solved :)
Nikita Araslanov retweeted
I'm super bored of LLM-stylized papers/blogs. I think we all are?
So I asked Claude to explain its tells: web.mit.edu/phillipi/www/claโฆ
Let's avoid these! I'd much rather read *your* words, see *your* rickety figures, than face down yet another of Claude's anodized corporate veneers.
Nikita Araslanov retweeted
๐ The #3DV2026 Nectar Track is open for submissions!
1๏ธโฃ Spotlight Track: Showcase your strong paper from recent venues with an oral presentation + poster.
2๏ธโฃ Exploration Edge Track: Present preliminary ideas, research hurdles, or demos in a poster session.
More details: โฌ๏ธ
Nikita Araslanov retweeted
TMLR has faced a deluge of submissions, necessitating stricter desk rejection policies due to limited reviewer capacity
Co-EiC Nihar Shah reached out to authors of 10 papers slated for desk reject. Could they answer questions about their *own* submission?
medium.com/@TmlrOrg/asking-aโฆ
Nikita Araslanov retweeted
"A Mathematical Explanation of Transformers" is a recent paper that develops a rigorous mathematical framework for understanding the architecture behind Transformers and large language models.
It interprets the Transformer as a discretization of a continuous integro-differential equation, with self-attention represented as a non-local integral operator, layer normalization as a projection onto a constrained set, and feedforward layers and activation functions incorporated into the same mathematical framework, then uses operator splitting and numerical discretization to recover the standard Transformer architecture and extend the formulation to multi-head attention, Vision Transformers, and convolutional Transformers.
I've already shared several resources on the mathematics behind neural networks, Transformers, and LLMs, but there always seems to be something new and interesting to explore in this area.
arxiv.org/pdf/2510.03989
Nikita Araslanov retweeted
We are seeing a new trend in submissions to @arxiv (and presumably to conferences and journals): Authors submitting papers whose contents they likely do not understand. 1/
Nikita Araslanov retweeted
MIT published a brutally honest report on what AI is doing to students.
A committee of professors and students spent five months studying how AI changed learning on campus, and the findings read like a warning to every university on the planet.
Study groups are disappearing. Office hours are emptying out. Problem sets and take-home exams no longer prove anything, because AI can produce credible solutions to almost any written assignment in the undergraduate curriculum. Students who lean on chatbots lose mastery and confidence, and some slip into what the report calls cognitive surrender, reaching for AI at the first hint of struggle.
The numbers are rough. 46 percent of surveyed MIT undergrads use LLMs daily. 90 percent worry about their own overreliance. Undergrads who feel AI makes them replaceable now outnumber those who feel it makes them capable.
The committee's answer surprised me. They refused to fight AI with surveillance. The report calls AI detectors unreliable, says lockdown browsers feel like spying, and warns that policing students builds a classroom atmosphere of mutual distrust.
Instead, MIT wants to rebuild education around the things AI can't replace. That means oral exams, semester portfolios, in-person project work, and a required social component in every subject. The report even floats the idea of rethinking grades entirely, since without a GPA to optimize, much of the incentive to cheat with AI evaporates.
The committee warns professors against replacing undergrad research assistants with AI agents just because they're cheaper, because a university exists to grow people, not output.
The most famous tech school on earth admitted the machines broke its way of teaching. Its answer is more humans, not more software.
Nikita Araslanov retweeted
๐ข [CVPRโ26] Can we learn to detect, segment, and track every object in a video without human supervision?ย
Yes, we introduce VideoCUPS, the first unsupervised video panoptic segmentation (VPS) method: 1. Get pseudo-labels from monocular videos. 2. Train a VPS model on them. โ at Denver, CO
Nikita Araslanov retweeted
Interested in efficient, high-resolution, temporally consistent features? Join our CVPR'26 Oral next
Saturday June 6 - Afternoon (14:00 - 15:15 UTC-6) ๐ธ
In-context learning suggests that a model has learned versatile representations. What if we use in-context learning itself as a training task for visual representations?
๐ฃ Introducing ๐๐๐๐: ๐๐ถ๐ป๐ฒ๐ฎ๐ฟ ๐๐ป-๐๐ผ๐ป๐๐ฒ๐
๐ ๐๐ฒ๐ฎ๐ฟ๐ป๐ถ๐ป๐ด โจ @CVPR 2026 Oral โจ
๐๐๐๐ trains on videos without manual annotation.
Key idea: An optimal linear mapping that predicts dense cues (e.g. depth, flow), estimated on one video frame, should also predict the corresponding cues of another frame from the same video.
This yields compelling results on dense vision tasks: video object segmentation, (zero-shot) semantic segmentation and surface normal estimation.
Paper, code, models and demo: lila-pixels.github.io
Joint work with @ma_sundermeyer, Hidenobu Matsuki, David Joseph Tan and @fedassa (and special thanks to David and Federico for hosting my research visit at Google).
#cvpr2026 @Google @MunichCenterML @tumcvg @TU_Muenchen