Kashyap Chitta's Avatar

Kashyap Chitta

@kashyap7x.bsky.social

kashyap7x.github.io Postdoc at NVIDIA. Previously at the University of TΓΌbingen and CMU. Robot Learning, Autonomous Driving.

2,159 Followers  |  599 Following  |  61 Posts  |  Joined: 16.11.2024  |  2.0659

Latest posts by kashyap7x.bsky.social on Bluesky

Preview
Percepta | A General Catalyst Transformation Company Transforming critical institutions using applied AI. Let's harness the frontier.

We're finally out of stealth: percepta.ai
We're a research / engineering team working together in industries like health and logistics to ship ML tools that drastically improve productivity. If you're interested in ML and RL work that matters, come join us πŸ˜€

02.10.2025 15:35 β€” πŸ‘ 118    πŸ” 18    πŸ’¬ 7    πŸ“Œ 2

We've hired some *fantastic* researchers but our startup is still looking for 2-3 more people with skills in ML/RL/LLMs. If you'd like to work on some transformative applied problems, hit me up. We'll be launching publicly soon too...

23.09.2025 17:31 β€” πŸ‘ 37    πŸ” 8    πŸ’¬ 0    πŸ“Œ 0
NVIDIA 2026 Internships: PhD Autonomous Vehicles Research - US By submitting your resume, you’re expressing interest in one of our 2026 Autonomous Vehicles focused Research Internships. We’ll review resumes on an ongoing basis, and a recruiter may reach out if yo...

We're recruiting our next batch of interns. Come join the team! nvidia.wd5.myworkdayjobs.com/en-US/NVIDIA...

18.09.2025 19:39 β€” πŸ‘ 6    πŸ” 1    πŸ’¬ 0    πŸ“Œ 0
How AI Taught Itself to See [DINOv3]
YouTube video by Jia-Bin Huang How AI Taught Itself to See [DINOv3]

How AI Taught Itself to See

Self-supervised learning is fascinating! How can AI learn from images only without labels?

In this video, we’ll build the method from first principles and uncover the key ideas behind CLIP, MAE, SimCLR, and DINO (v1–v3).

Video link: youtu.be/oGTasd3cliM

16.09.2025 23:13 β€” πŸ‘ 13    πŸ” 2    πŸ’¬ 0    πŸ“Œ 0

Putting the final touches on your submission? Just a few days left to enter the @iccv.bsky.social NAVSIM Challenge! Deadline: September 20. $8K in prizes and several travel grants are on the line!

15.09.2025 14:02 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
CS208: Canon of Computer Science, Spring 2010

Found this marvelous little course full of readings to trace the evolution of computer science and its canonical ideas
graphics.stanford.edu/courses/cs20...

14.09.2025 19:58 β€” πŸ‘ 46    πŸ” 4    πŸ’¬ 2    πŸ“Œ 0
Post image

Our Principal Investigators Antonio Orvieto, Celestine Mendler-DΓΌnner, Maximilian Dax, Rediet Abebe, @shiweiliu.bsky.social, T. Konstantin Rusch, and @wielandbrendel.bsky.social are looking for motivated students interested in internships.
Apply here: docs.google.com/forms/d/e/1F...

11.09.2025 09:53 β€” πŸ‘ 2    πŸ” 2    πŸ’¬ 1    πŸ“Œ 1
Post image

Announcing the @iccv.bsky.social NAVSIM Challenge! What's new? We're testing not only on real recordings, but also perturbed futures generated from the real ones via pseudo-simulation! $8K in prizes + several $1.5k travel grants. Submit by September 20! opendrivelab.com/challenge2025/ πŸ§΅πŸ‘‡

01.09.2025 09:14 β€” πŸ‘ 14    πŸ” 9    πŸ’¬ 1    πŸ“Œ 1
Preview
GitHub - autonomousvision/navsim: [CoRL '25] Pseudo-Simulation for Autonomous Driving; [NeurIPS '24] NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarking [CoRL '25] Pseudo-Simulation for Autonomous Driving; [NeurIPS '24] NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarking - autonomousvision/navsim

Your score depends on:
1️⃣ Performance on real frames
2️⃣ Performance on "nearest" perturbed futures to your model's past outputs,
making this the *first* challenge testing robustness against causal confusion and compounding errors on real driving data πŸš—

Get started here: github.com/autonomousvi...

01.09.2025 09:14 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image

Announcing the @iccv.bsky.social NAVSIM Challenge! What's new? We're testing not only on real recordings, but also perturbed futures generated from the real ones via pseudo-simulation! $8K in prizes + several $1.5k travel grants. Submit by September 20! opendrivelab.com/challenge2025/ πŸ§΅πŸ‘‡

01.09.2025 09:14 β€” πŸ‘ 14    πŸ” 9    πŸ’¬ 1    πŸ“Œ 1
Preview
ELLIS PhD Program: Call for Applications 2025 The ELLIS mission is to create a diverse European network that promotes research excellence and advances breakthroughs in AI, as well as a pan-European PhD program to educate the next generation of AI researchers. ELLIS also aims to boost economic growt...

πŸŽ“ Interested in a #PhD in machine learning or #AI? The ELLIS PhD Program connects top students with leading researchers across Europe. The application portal opens on Oct 1st. Curious? Join our info session on the same day. Get all the info πŸ‘‰

29.08.2025 07:14 β€” πŸ‘ 16    πŸ” 13    πŸ’¬ 0    πŸ“Œ 2

Let's push for the obvious solution: Dear @neuripsconf.bsky.social ! Allow authors to present accepted papers at EurIPS instead of NeurIPS rather than just additionally. Likely, at least 500 papers would move to Copenhagen, problem solved.

28.08.2025 19:19 β€” πŸ‘ 56    πŸ” 15    πŸ’¬ 2    πŸ“Œ 0

πŸš€ Introducing our new paper, MDPO: Overcoming the Training-Inference Divide of Masked Diffusion Language Models.

πŸ“„ Paper: www.scholar-inbox.com/papers/He202...
arxiv.org/pdf/2508.13148
πŸ’» Code: github.com/autonomousvi...
🌐 Project Page: cli212.github.io/MDPO/

20.08.2025 09:30 β€” πŸ‘ 12    πŸ” 8    πŸ’¬ 1    πŸ“Œ 2
Post image

Introducing DINOv3 πŸ¦•πŸ¦•πŸ¦•

A SotA-enabling vision foundation model, trained with pure self-supervised learning (SSL) at scale.
High quality dense features, combining unprecedented semantic and geometric scene understanding.

Three reasons why this mattersπŸ‘‡

14.08.2025 18:50 β€” πŸ‘ 24    πŸ” 8    πŸ’¬ 2    πŸ“Œ 2
Preview
GitHub - autonomousvision/CaRL: [CoRL 2025] CaRL: Learning Scalable Planning Policies with Simple Rewards [CoRL 2025] CaRL: Learning Scalable Planning Policies with Simple Rewards - autonomousvision/CaRL

Our paper "CaRL: Learning Scalable Planning Policies with Simple Rewards" has been accepted to the Conference on Robot Learning (CoRL 2025).
See you in Seoul at the end of September.

Code & Paper:
github.com/autonomousvi...

08.08.2025 11:38 β€” πŸ‘ 3    πŸ” 1    πŸ’¬ 0    πŸ“Œ 0

At TRI’s newest division: Automated Driving Advanced Development, we’re building a clean-slate, end-to-end autonomy stack. We're hiring, with open roles in learning, infra, and validation: www.tri.global/careers#open...

05.08.2025 22:57 β€” πŸ‘ 4    πŸ” 1    πŸ’¬ 0    πŸ“Œ 0
Post image

1/ Can open-data models beat DINOv2? Today we release Franca, a fully open-sourced vision foundation model. Franca with ViT-G backbone matches (and often beats) proprietary models like SigLIPv2, CLIP, DINOv2 on various benchmarks setting a new standard for open-source research.

21.07.2025 14:47 β€” πŸ‘ 84    πŸ” 21    πŸ’¬ 2    πŸ“Œ 3
Post image

Adding a nice way to visualize the PPO objective to the rlhf book.
Core for policy-gradient is L is proportional to R*A (R=policy ratio, A = advantage).
PPO makes good actions more likely, up to a point.
PPO makes bad actions less likely, up to a point.

19.07.2025 23:10 β€” πŸ‘ 12    πŸ” 2    πŸ’¬ 2    πŸ“Œ 0
Post image

What if the cherry was actually cake? (source: arxiv.org/abs/2506.08007)

20.07.2025 10:18 β€” πŸ‘ 5    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

🚨Job Alert
W2 (TT W3) Professorship in Computer Science "AI for People & Society"
@saarland-informatics-campus.de/@uni-saarland.de is looking to appoint an outstanding individual in the field of AI for people and society who has made significant contributions in one or more of the following areas:

18.07.2025 07:11 β€” πŸ‘ 14    πŸ” 18    πŸ’¬ 1    πŸ“Œ 0
Post image

EurIPS is coming! πŸ“£ Mark your calendar for Dec. 2-7, 2025 in Copenhagen πŸ“…

EurIPS is a community-organized conference where you can present accepted NeurIPS 2025 papers, endorsed by @neuripsconf.bsky.social and @nordicair.bsky.social and is co-developed by @ellis.eu

eurips.cc

16.07.2025 22:00 β€” πŸ‘ 139    πŸ” 70    πŸ’¬ 2    πŸ“Œ 17
Post image

In an era of billion-parameter models everywhere, it's incredibly refreshing to see how a fundamental question can be formulated and solved with simple, beautiful math.

- How should we orient a solar panel β˜€οΈπŸ”‹? -

Zero AI! If you enjoy math, you'll love this!

Video: www.youtube.com/watch?v=ZKzL...

16.07.2025 14:25 β€” πŸ‘ 9    πŸ” 2    πŸ’¬ 1    πŸ“Œ 0
Post image

πŸ“’ Present your NeurIPS paper in Europe!

Join EurIPS 2025 + ELLIS UnConference in Copenhagen for in-person talks, posters, workshops and more. Registration opens soon; save the date:

πŸ“… Dec 2–7, 2025
πŸ“ Copenhagen πŸ‡©πŸ‡°
πŸ”—eurips.cc

#EurIPS
@euripsconf.bsky.social‬

16.07.2025 23:00 β€” πŸ‘ 61    πŸ” 17    πŸ’¬ 2    πŸ“Œ 3
Post image Post image

I really like this paper on relative positional encodings using projective geometry for multi-view transformers, by Li et al. (Berkeley/Nvidia/HKU).

It is elegant: in special situations, it defaults to known baselines like GTA (if identity intrinsics) and RoPE (same cam).

arxiv.org/abs/2507.10496

15.07.2025 14:40 β€” πŸ‘ 23    πŸ” 3    πŸ’¬ 0    πŸ“Œ 1
Preview
GitHub - autonomousvision/CaRL: [ArXiv 2025] CaRL: Learning Scalable Planning Policies with Simple Rewards [ArXiv 2025] CaRL: Learning Scalable Planning Policies with Simple Rewards - autonomousvision/CaRL

We have released the code for our work, CaRL: Learning Scalable Planning Policies with Simple Rewards.

The repository contains the first public code base for training RL agents with the CARLA leaderboard 2.0 and nuPlan.

github.com/autonomousvi...

15.07.2025 16:05 β€” πŸ‘ 20    πŸ” 7    πŸ’¬ 0    πŸ“Œ 2
CaRL: Learning Scalable Planning Policies with Simple Rewards
YouTube video by Daniel Dauner CaRL: Learning Scalable Planning Policies with Simple Rewards

In case you find it as relaxing as we do: Here is a 2h+ video of our autonomous RL driving agent CaRL in action! @danieldauner.bsky.social @bernhard-jaeger.bsky.social @kashyap7x.bsky.social
youtube.com/watch?v=_god...

15.07.2025 06:17 β€” πŸ‘ 21    πŸ” 5    πŸ’¬ 0    πŸ“Œ 1

🏟️But we’re not done yet - our workshop continues at #ICCV2025! And the challenge moves forward too, with more prizes and exciting updates. opendrivelab.com/challenge2025/

13.07.2025 16:39 β€” πŸ‘ 3    πŸ” 2    πŸ’¬ 0    πŸ“Œ 0

Do you also plan to open-source this continuous token version? As I understand, the current release is for a discrete-tokenized model set, right?

10.07.2025 23:37 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
Post image Post image Post image Post image

Our @matthieucord.bsky.social talked on Driving through Generative video Pretraining: VaVIM-VaVAM.
He talked on how we boosted speed & perf. by reducing number of tokens and switching from discrete to continuous tokens

10.07.2025 23:16 β€” πŸ‘ 4    πŸ” 1    πŸ’¬ 2    πŸ“Œ 0
Post image

πŸš— Excited to announce our RealADSim Workshop & Challenges @ ICCV 2025 β€” join us!
πŸ† Over $40,000 in prizes
πŸŒ† Competition 1: Extrapolated Novel View Synthesis
🚘 Competition 2: Closed-loop Driving in a 3DGS-based Simulator
🎀 Featuring great speakers
πŸ”— realadsim.github.io/2025/

30.06.2025 14:09 β€” πŸ‘ 6    πŸ” 2    πŸ’¬ 0    πŸ“Œ 1

@kashyap7x is following 20 prominent accounts