Priyam Parashar

I enjoy music, reading for pain and leisure, and writing stuff once in a while. Employed as a roboticist/MLE @ Waymo, all thoughts are my own.

[Priyam Parashar @priyam8parashar]

Text-as-knowledge for AI agents and Obsidian — now with CLI! — just make sense.

[Obsidian @obsdmd]

Anything you can do in Obsidian you can do from the command line.

Obsidian CLI is now available in 1.12 (early access).

[Sam Altman @sama]

hot ai summer lfg

[Akash Sharma @akashshrm02]

Robots need touch for human-like hands to reach the goal of general manipulation. However, approaches today don’t use tactile sensing or use specific architectures per tactile task.

Can 1 model improve many tactile tasks?
🌟Introducing Sparsh-skin: https://t.co/DgTq9OPMap

1/6

[AI at Meta @AIatMeta]

Introducing Meta Locate 3D: a model for accurate object localization in 3D environments.

Learn how Meta Locate 3D can help robots accurately understand their surroundings and interact more naturally with humans.

You can download the model and dataset, read our research paper, and even try a demo! https://t.co/GgJEPXTH8W

[Priyam Parashar @priyam8parashar]

Here I’m talking about our recent work on PARTNR here. Check out the video and find our paper and code at: https://t.co/sRUwt3DEcd, https://t.co/mOhOqbVlhU

[AI at Meta @AIatMeta]

A closer look at the evolution of the research from Meta FAIR that led to our latest open source work for human-robot collaboration.

Details on Meta PARTNR ➡️ https://t.co/CgPcQ8OpTy

[Priyam Parashar @priyam8parashar]

Our work on PARTNR released on 10th anniversary of research @ FAIR Paris @ Meta! What fun to work with amazing people studying LLM applications to human-robot collaboration. Check out the blog at the link in the post.

Paper: https://t.co/sRUwt3DEcd
Code: https://t.co/mOhOqbVlhU

[Joelle Pineau @jpineau1]

As part of our work towards Advanced Machine Intelligence I’m excited to share a few new open source releases and milestones from our team at Meta FAIR!

https://t.co/knwckdltG9

[So Yeon (Tiffany) Min @SoYeonTiffMin]

We introduce Situated Instruction Following (SIF), to appear in ECCV 2024! There is inherent underspecification in instructions when humans act as they speak. SIF is addresses these dynamic, temporally evolving intent of instructions in the context of physical human actions.(1/7)

[Priyam Parashar @priyam8parashar]

@keerthanpg I am sure a lot of other literature has contributed but what surfaces to mind is prominent books around local creation of meaning and universal lack of it, i.e. absurdism.
1. Man's search for meaning by V. Frankl (former)
2. The Stranger by A. Camus (latter)

[Keerthana Gopalakrishnan @keerthanpg]

One of the things I didn’t appreciate as a younger researcher but that which I do now is how important it is to make things up and tell moving stories, in ways that appeal to people’s imaginations.

[Jim Fan @DrJimFan]

AI winter? No. Even if GPT-5 plateaus. Robotics hasn’t even started to scale yet.

Embodied intelligence in the physical world will be a powerhouse for economic value. Friendly reminder to everyone that LLM is not all of AI. It is just one piece of a bigger puzzle.

[Priyam Parashar @priyam8parashar]

@langchain @memdotai mem it

[Andrii Zadaianchuk 🇺🇦 @ZadaianchukML]

📢 We are excited to introduce Thomas Kipf (@tkipf), Animesh Garg (@animesh_garg), and Sungjin Ahn (@SungjinAhn_) as the first three speakers of the Causal and Object-Centric Representations for Robotics (CORR) CVPR workshop.

https://t.co/uSTAKjihYr
🧵 1/5

[Priyam Parashar @priyam8parashar]

@saramagliacane @AmlabUva @UvA_Amsterdam @MITIBMLab @RAEngNews @Oxford_VGG @UniofOxford @RaresAmbrus leads the Analysis-By-Synthesis (ABS) team at @ToyotaResearch. His research interests lie at the intersection of robotics, computer vision and machine learning, with an emphasis on self-supervised learning for 3D perception.
---
Deadline for submissions: 04/28!
🧵4/4

[Priyam Parashar @priyam8parashar]

Introducing our speakers for Causal and Object-centric Representations for Robotics (CORR) CVPR Workshop. We are so excited to have Sara Magliacane (@saramagliacane), Joao F. Henriques and Rares Ambrus (@RaresAmbrus) at our workshop!!
🧵1/4

[Andrii Zadaianchuk 🇺🇦 @ZadaianchukML]

📢 We are excited to introduce Thomas Kipf (@tkipf), Animesh Garg (@animesh_garg), and Sungjin Ahn (@SungjinAhn_) as the first three speakers of the Causal and Object-Centric Representations for Robotics (CORR) CVPR workshop.

https://t.co/uSTAKjihYr
🧵 1/5

[Priyam Parashar @priyam8parashar]

@saramagliacane @AmlabUva @UvA_Amsterdam @MITIBMLab Dr. João Henriques is a Research Fellow at @RAEngNews, and a lead PI at the @Oxford_VGG-@UniofOxford. His research focuses on computer vision and deep learning, with the goal of making machines more perceptive, intelligent and capable of helping people.
🧵3/4

[Priyam Parashar @priyam8parashar]

@profembwall This is amazing. Congratulations!

[Priyam Parashar @priyam8parashar]

@deedydas More reported incidents do not necessarily mean objective rise in underlying causes. Confounding reasons: (a) relatively new collation and digitisation of data, (b) larger health care coverage, (c) rising economy enabling more healthcare visits.

[Chris Paxton @chris_j_paxton]

Robotics foundation model is a term you will hear a lot over the next year

[Priyam Parashar @priyam8parashar]

Also this is my way of telling myself and everyone in the void: The Angel of Indian Lake by @SGJ72 is out!! My plans for this weekend are set.

[Priyam Parashar @priyam8parashar]

This was such a great episode: https://t.co/iy9hlN5nSd

@mangiotto talks so deeply about Kaiju genre and its creation. It’s well worth a listen to if you’ve ever loved Godzilla in any way. Ending shouts out to @SGJ72 and his amazing Indian Lake Trilogy.

10/10.