We recently pushed an update to this paper. Usually, updates don't justify a post, but this one is exceptionally contentful -> 🧵
tldr: all the findings are stronger, and the behaviors are super cool!
arxiv.org/abs/2410.05362
I am looking for a postdoc. A serious-looking call coming soon, but this is to get it going. Topics include (but not limited to): LLMs (🫢!), multimodal LLMs, interaction+learning, RL, intersection with cogsci, ... see our work to get an idea:
yoavartzi.com/pubs
Plz RT 🙏
.@Cornell is recruiting for multiple postdoctoral positions in AI as part of two programs: Empire AI Fellows and Foundational AI Fellows. Positions are available in NYC and Ithaca.
Deadline for full consideration is Nov 20, 2025!
academicjobsonline.org/ajo/jobs/30971
It's now public! My postdoc call is for the inaugural postdoc as part of this $10.5M gift for a new AI fellows program at Cornell. There's a lot more in this program, so more exciting things to happen here real soon!
news.cornell.edu/stories/2025/0…
Application: forms.gle/tiydAChgV1wLcQ…
I am looking for a postdoc. A serious-looking call coming soon, but this is to get it going. Topics include (but not limited to): LLMs (🫢!), multimodal LLMs, interaction+learning, RL, intersection with cogsci, ... see our work to get an idea:
yoavartzi.com/pubs
Plz RT 🙏
The video for this talk is now online:
youtube.com/watch?v=ozHk_b…
Paper here:
Pre-training Limited Memory Language Models with Internal and External Knowledge
Linxi Zhao, Sofian Zalouk, Christian K. Belardi, Justin Lovelace, Jin Peng Zhou, Ryan Thomas Noonan, Dongyoung Go, Kilian
Yoav Artzi (@yoavartzi) on Limited Memory Language Models (LMLMs)
-- disentangle knowledge and competency
-- RAG focuses on performance, and still rely on LLMs that run at data centers. It's mostly a post-training technique.
-- LMLMs focus is on running LLMs on your own device,
Introducing our new mechanistic interpretability paradigm: Proctology of Language Models! It's based on LLMs judging each other's activations to be bad/good/super. The names is inspired from dog butt-sniffing
introducing our new interpretability research paradigm, Cardiology of Language Models! it is based on a method we call the "stethoscope", where we train a linear classifier to discriminate between the LLM hidden states that represent a concept and those that do not!