We are getting closer to have agents operating in the real physical world. However, can we trust frontier models to make embodied decisions ๐ฎ aligned with human norms ๐ฉโโ๏ธ ?
With EgoNormia, a 1.8k ego-centric video ๐ฅฝ QA benchmark, we show that this is surprisingly challenging!
04.03.2025 04:32 โ ๐ 23 ๐ 9 ๐ฌ 1 ๐ 1
How many (checks calendar) decades do people keep around backups of data from their thesis? Am I a digital hoarder?
05.01.2025 05:00 โ ๐ 10 ๐ 0 ๐ฌ 4 ๐ 0
Recently, papers have been published in prestigious journals (Nature Human Behaviour, PNAS) claiming that large language models (e.g., ChatGPT) solve the "false belief" task (a task requiring Theory of Mind abilities).
What is the false belief task? ->
17.12.2024 08:36 โ ๐ 6 ๐ 2 ๐ฌ 1 ๐ 0
I think Bisky :)
25.11.2024 20:12 โ ๐ 2 ๐ 0 ๐ฌ 0 ๐ 0
When I first started reading papers in ~2007 we would wait for someone who attended the conference to bring back the conference booklet and tell us what to read. Weโd read them. And then spend the next three months reading old papers or working cuz ๐คท It was a great way to grow up.
24.11.2024 14:52 โ ๐ 9 ๐ 0 ๐ฌ 0 ๐ 0
๐ Time is a weird thing :) maybe I should try and convince the department to force you to give a talk sometime โ though ideally with @spandanagella.bsky.social too ;)
24.11.2024 02:30 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0
Just use your own domain name?
24.11.2024 02:13 โ ๐ 0 ๐ 0 ๐ฌ 1 ๐ 0
This article really spoke to me; all the science I've enjoyed and that I thought came out well has been done with a colleague that I was talking to every day and almost every couple of hours
17.11.2024 14:32 โ ๐ 42 ๐ 3 ๐ฌ 1 ๐ 0
Hello, Computational linguistics/NLP world in Bluesky! We're creating the same accounts on other social media platforms in Bluesky! #NLProc
14.11.2024 00:17 โ ๐ 133 ๐ 31 ๐ฌ 4 ๐ 5
I am trying to create a robotics and ai starter pack on bluesky: go.bsky.app/DfAoaJ1
Very incomplete please comment with suggestions (or just if you're missing and want to be added!)
11.11.2024 15:01 โ ๐ 111 ๐ 38 ๐ฌ 77 ๐ 4
3. How to Train Your Fact Verifier: Knowledge Transfer with Multimodal Open Models
^^ includes @skgabrie.bsky.social who is just starting up her lab at UCLA!
10.11.2024 18:34 โ ๐ 4 ๐ 0 ๐ฌ 0 ๐ 0
2. Gradient Localization Improves Lifelong Pretraining of Language Models
TL;DR - Gradient norms tell you where your knowledge is stored and if it conflicts with what you already know.
10.11.2024 18:34 โ ๐ 2 ๐ 0 ๐ฌ 1 ๐ 0
#EMNLP2024
1. Tools Fail: Detecting Silent Errors in Faulty Tools
Are you using tools with your LLMs? Are you assuming your tools are perfect? Assuming the LLM can just handle any errors for you? ๐ฌ
Dangerโฆ ๐จ Models trust tools over their own โknowledgeโ even for simple and well trained cases.
10.11.2024 18:34 โ ๐ 19 ๐ 0 ๐ฌ 1 ๐ 0
TRI Women and Allies breakfast
A robot standoff. One with wheels and one with legs.
Su debugging her robot teleop system
Vidhi presenting her work on robot audio
Hi from CoRL ๐
08.11.2024 10:27 โ ๐ 15 ๐ 2 ๐ฌ 1 ๐ 0
Assistant Professor at @cs.ubc.caโฌ and โช@vectorinstitute.aiโฌ working on Natural Language Processing. Book: https://lostinautomatictranslation.com/
Postdoc @UNC working on NLP, AI, and computational linguistics. Formerly PhD student @JHU and undergrad @McGill
esteng.github.io
Assistant Prof. at Georgia Tech | NVIDIA AI | Making robots smarter
Associate Professor at #MIT, SPARK Lab Director, Roboticist, interested in how machines see and understand the world
lucacarlone.mit.edu
Assistant Professor in Computer Science at UofT.
wrote a book called "in this economy?" | chair of the federal reserve | writing and youtube @ http://kyla.substack.com
In-depth, independent reporting to better understand the world, now on Bluesky. News tips? Share them here: http://nyti.ms/2FVHq9v
Official account for #Taskmaster.
Watch full episodes on YouTube and Channel4.com.
http://theverge.com covers life in the future.
At wired.com where tomorrow is realized || Sign up for our newsletters: https://wrd.cm/newsletters
Find our WIRED journalists here: https://bsky.app/starter-pack/couts.bsky.social/3l6vez3xaus27
The first and only satirical women's magazine. Constantly moving to a new social platform.
Husband, dad, veteran, writer, and proud Midwesterner. 19th US Secretary of Transportation and former Mayor of South Bend.
She/Her ๐
Clownery is the best medicine
Bluesky is where I post my nonsenseโจ
PBSโs editorial independence is central to our work and will never change. We produce trustworthy content that features unbiased reporting.
http://www.pbs.org
We are home to PBS News Hour (ranked the most credible and objective TV news show), PBS News Weekend and @washingtonweekpbs.bsky.social.
Donate now to support our work: https://bit.ly/3IXO4xW
More: linktr.ee/pbsnews
Proudly serving the people of New Jersey in the U.S. Senate.
Computation Cognition Learning
PhD@cmu
The Computer Science Department's mission has remained steadfast: to lead in computer science research and education that has real-world impact โ to push the frontiers of the field and produce the next generations leaders.
Cybersecurity professional on sabbatical in Sweden. Aviation nerd. SF enthusiast. Warhammer 40k painter and only occasional player.