arya @mimikdev

some people think it's an insult to the glory of their sickness to get well prev intern @aiatmeta @netflix @amazon avalon - they Joined August 2023

Tweets

131
Followers

2K
Following

279
Likes

725

will brown @willccbb

a week ago

if you’re a professor teaching about LLM RL this semester + considering doing any sort of hands-on lessons about RL environments/agentic RL, hit me up, would love to chat :) this stuff is now at the accessibility level where students can easily play with it

26 36 652 48K 317

Tim Hua 🇺🇦 @Tim_Hua_

a month ago

AIs can have secret “backdoors:” Can we uncover them? Maybe! We study models that misbehave in an unknown situation, and successfully reverse-engineer the situation in simple settings. We weren't sure this was possible! We hope to inspire future work in more realistic settings.

3 9 80 16K 42

Download Image

arya @mimikdev

4 months ago

"ai evangelist"

machine yearning engineer @confusionm8trix

4 months ago

"ai evangelist"

2K 6K 157K 18.1M 20K

0 0 22 7K 1

arya @mimikdev

4 months ago

honestly a pretty crazy journey coming from being a neuroscience student 2 years ago.. gonna build something to share my process and learnings with you all :)

arya @mimikdev

4 months ago

honestly a pretty crazy journey coming from being a neuroscience student 2 years ago.. gonna build something to share my process and learnings with you all :)

10 1 289 22K 30

4 0 37 6K 6

arya @mimikdev

4 months ago

joining @Meta this summer to work on ML Automation :) been a bit MIA recently but just wanted to give an update

10 1 289 22K 30

arya @mimikdev

8 months ago

been locked tf in recently also trying to get more into some of the high value content on youtube like stanfords cs153 guest speaker videos

1 0 20 3K 5

Alice Rigg @woog09

8 months ago

New paper from @norabelrose and I. We show how mech interp can be done on generic relu networks--a feat previously understood to be intractable. Rather than enumerate over polytopes we OLS regress on max entropy inputs, deriving guarantees on model perf. arxiv.org/abs/2502.01032

Nora Belrose @norabelrose

8 months ago

5 32 300 31K 200

Download Image

0 9 43 5K 10

Khush Gupta @khushgx

10 months ago

come check out our poster at the ATTRIB Workshop (west room 207)!

3 3 19 3K 1

Download Image

arya @mimikdev

10 months ago

One of the things that excited me the most about pursuing a PhD in the future is finally being qualified to mentor bright undergraduate students who want a career in research. There's something beautiful about helping someone grow into who they want to be.

2 0 37 2K 5

arya @mimikdev

11 months ago

reminder to donate to wikipedia if you are fortunate enough to be able to do so :)

0 1 14 1K 0

Abhilasha Ravichander @lasha_nlp

11 months ago

✨I’m on the faculty job market for 2024-2025! ✨ My research focuses on advancing Responsible AI—enhancing factuality, robustness, and transparency in AI systems. I’m at #EMNLP2024 this week🌴 and would love to chat about research and hear any advice!

1 46 224 20K 9

arya @mimikdev

11 months ago

you learn so many new things when trying to explore the viability of a new research idea its actually amazing

0 1 18 2K 0

Valentina Pyatkin @valentina__py

11 months ago

📈New paper on implicit language and context! She bought the largest pumpkin? - Largest pumpkin out of what? All pumpkins in the store? Out of all pumpkins bought by her friends? In the world? Superlatives are (often) ambiguous and their interpretation is extremely context…

1 18 125 9K 35

Download Image

arya @mimikdev

12 months ago

going down the rabbit hole of papers and blogs to inform myself, finding some rly cool stuff. today gonna setup a notion for technical blogging in general + setup the repo and outlining next steps :]

arya @mimikdev

12 months ago

going down the rabbit hole of papers and blogs to inform myself, finding some rly cool stuff. today gonna setup a notion for technical blogging in general + setup the repo and outlining next steps :]