Andrew LampinenDec 8, 2023

Very excited to head to NeurIPS! Feel free to reach out if you want to chat about any of our recent work on LMs, agents, interpretability, representational alignment, etc. You can find me:

At the poster for our work on "Passive learning of active causal strategies in agents and language models"
https://sigmoid.social/@lampinen/110434383859776741
Tue 12 Dec 5:15 p.m. CST — 7:15 p.m. CST
Great Hall & Hall B1+B2 (level 1) #825
https://nips.cc/virtual/2023/poster/72481
1/3
#Neurips #neurips2023

Andrew Lampinen (@[email protected])

What can be learned about causality and experimentation from passive data? What could language models learn from simply passively imitating text? We explore these questions in our new paper: “Passive learning of active causal strategies in agents and language models” https://arxiv.org/abs/2305.16183 Thread: 1/7 (x-post from https://twitter.com/AndrewLampinen/status/1662006807693336582?s=20) #llm #RL #causality #experiment #deeplearning #languagemodels #nlp #nlproc

Sigmoid Social
Show threadAndrew Lampinen
At the poster for
@lukasmut's paper "Improving neural network representations using human similarity judgments"
https://sigmoid.social/@lukasmut/110524246202395409
Wed 13 Dec 10:45 a.m. CST — 12:45 p.m. CST
Great Hall & Hall B1+B2 (level 1) #300
https://nips.cc/virtual/2023/poster/71848
And while you're in the Wednesday morning session, I can't make it due to the clash, but check out
@cogscikid
's great paper on "Combining behaviors with the successor feature keyboard"
https://twitter.com/cogscikid/status/1731683266942423318?t=b8CTL79Z6b01B8dyrypn_Q&s=19
#1913
https://nips.cc/virtual/2023/poster/72199
2/4
Lukas Muttenthaler (@[email protected])

🚨Beep beep 🚨 Have you ever been wondering about how to use human similarity judgments for improving neural network representations? 🧠 We have something for you! We found a linear transform that improves both representational alignment and downstream task performance! 🦾 https://arxiv.org/abs/2306.04507

Sigmoid Social
Dec 8, 2023 at 8:04pmWeb
Show threadAndrew LampinenDec 8, 2023
Finally, I'll be a panelist and speaker at the UniReps workshop on Friday: unireps.org
where I'll be discussing and speaking about some of the issues we've been thinking about in our recent work on representational alignment:
https://sigmoid.social/@lampinen/111347414419226116
3/4
Andrew Lampinen (@[email protected])

Attached: 1 image What is representational alignment? How can we use it to study or improve intelligent systems? What challenges might we face? In a new paper (arxiv.org/abs/2310.13018 ), we describe a framework that attempts to unify ideas from cognitive science, neuroscience and AI to address these questions. This paper is very much a work in progress, so feedback welcome! And thanks to all the contributors, and particularly the first authors Ilia & @lukasmut for their hard work!

Sigmoid Social
Show threadAndrew LampinenDec 8, 2023
On a more personal note, this trip also marks the end of my time living in London — will miss this city and living around the corner from this view, but excited for what's to come! 4/4