Kumar Ashutosh

CS PhD student at @UTAustin | Prev - @AIatMeta, @iitbombay.

[Kumar Ashutosh @ashutoshkr7]

Check out our latest work in video generation! 🚀
We introduce a simple human motion reward model to post-train video diffusion models for realistic human motion with correct dynamics and anatomy. 🧠🎥

Check out the project page for more videos!

Webpage: https://t.co/o8cgGnTrbO

[Rohit Girdhar @_rohitgirdhar_]

Excited to share our latest work on video generation!!

Video models, even state-of-the-art ones like Sora-2, struggle to generate complex human motion. Here’s a video of me doing a backflip 😂 (fwiw, my real-life backflips aren’t any better 😅)

Here's how to fix it 👇
1/5

[Sumith Kulal @sumith1896]

Diffusion Circle happening today @CVPR #CVPR2026

Upper Lobby F at 3:30pm, come join us!

[Georgios Pavlakos @geopavlakos]

Ashutosh @chargedneutron_ is presenting ExpertAF during this poster session! We are at poster #280. Come by to chat about it!

Project page: https://t.co/nz82YDcoPk

[Kumar Ashutosh @ashutoshkr7]

Check out our new work in enabling LLMs to see and hear _without_ any training.

Paper: https://t.co/EgWsw6Fz3O
Code: https://t.co/pnvbttCOmC

[Rohit Girdhar @_rohitgirdhar_]

Super excited to share some recent work that shows that pure, text-only LLMs, can see and hear without any training! Our approach, called "MILS", uses LLMs with off-the-shelf multimodal models, to caption images/videos/audio, improve image generation, style transfer, and more!

[Kumar Ashutosh @ashutoshkr7]

Check out VidOSC, to be presented at CVPR 2024 🎉

[Zihui (Sherry) Xue @sherryx90099597]

🌟Introducing our #CVPR2024 work on video object state changes

We present a novel open-world problem formulation, a new large-scale dataset, and a learning framework for temporally localizing different stages of an object state change

Explore more: https://t.co/Z63OdsDhna

[Kumar Ashutosh @ashutoshkr7]

Work done at UT Austin and @AIatMeta. Collaborators: @sherryx90099597, @TusharNagarajan and Kristen Grauman.

[Kumar Ashutosh @ashutoshkr7]

Introducing our #CVPR2024 (Highlight ✨) paper, "Detours for Navigating Instructional Videos."

Imagine watching an instructional video and having questions like, "Can I add carrots here?" Our novel VidDetours enable video detours that use user queries and prior viewing context.

[Kumar Ashutosh @ashutoshkr7]

We have created a weakly-supervised dataset and a manually annotated testing dataset for this new task.
Paper: https://t.co/oqD5No5in9
Project page: https://t.co/42AjSfi8j4
Code:
https://t.co/OT133ktDGe

[Kumar Ashutosh @ashutoshkr7]

🚀🚀

[#CVPR2026 @CVPR]

HUGE shoutout to our #CVPR2024 Outstanding Reviewers 🫡

[Kumar Ashutosh @ashutoshkr7]

We evaluate our method on several video datasets on step forecasting, step recognition and task recognition and see consistent gains over prior work.

Project page: https://t.co/fFzlpd7pc6

Please come by the poster session today (12/12 morning session) at #129.

[Kumar Ashutosh @ashutoshkr7]

Introducing our #NeurIPS23 paper on using video-mined task graphs for keystep recognition in instructional videos. 🎉

We first use videos on the internet to learn a task graph and then use the learned graph to regularize keystep recognition in instructional videos.

[Kumar Ashutosh @ashutoshkr7]

Webpage: https://t.co/UvsYf5Oh5C
Paper: https://t.co/OpztSbWqXY
Blog: https://t.co/wjgnEtxzz0

[Kumar Ashutosh @ashutoshkr7]

Exciting collaboration between 14 research institutions resulting in a first-of-its kind dataset pushing the frontiers in skill understanding. Covers cooking, sports, dance and health - thus enabling research in multiple domains and tasks. Glad to be a part of this effort!🎉

[AI at Meta @AIatMeta]

1️⃣ Ego-Exo4D
A new foundational dataset + benchmark suite to support research on video learning & multimodal perception, co-developed with 14 university partners.

Details ➡️ https://t.co/DmQXYvmFIN

Core to the work is videos of skilled human activities, simultaneously capturing both first-person "egocentric" + multiple “exocentric” views.

[Kumar Ashutosh @ashutoshkr7]

Exciting to have both squash and cricket in the LA Olympics 🥳

[The Olympic Games @Olympics]

IOC Session approves @LA28’s proposal for 5⃣ additional sports:

⚾Baseball/🥎softball, 🏏cricket, 🏈flag football, 🥍lacrosse and ⚫squash have been officially included as additional sports on the programme for the Olympic Games Los Angeles 2028. #LA28

[Lex Fridman @lexfridman]

Here's my conversation with Mark Zuckerberg, his 3rd time on the podcast, but this time we talked in the Metaverse as photorealistic avatars. This was one of the most incredible experiences of my life. It really felt like we were talking in-person, but we were miles apart 🤯 It's hard to put into words how awesome this was for someone like me who values the intimacy of in-person conversation. It gave me a glimpse of an exciting future with many new possibilities and fascinating questions about the nature of reality and human connection ❤

Timestamps:
0:00 - Introduction
0:52 - Metaverse
15:27 - Quest 3
30:16 - Nature of reality
34:54 - AI in the Metaverse
51:51 - Large language models
57:49 - Future of humanity

[Kumar Ashutosh @ashutoshkr7]

@kalpeshk2011 @MohitIyyer @andrewmccallum @HamedZamani @YejinChoinka @Google Congrats Kalpesh!

[Lex Fridman @lexfridman]

Beautiful 60-day timelapse of a bird building a nest and raising her kids until they fly away from home. This is the magic of life.

[Kumar Ashutosh @ashutoshkr7]

@MengTo Really liked your TL;DR! I also lived briefly in Japan and that spoiled me. Did you not consider moving to Japan? If no, what were the reasons?