🔥 New position piece! 🔥 In this paper we lay out our vision for AI Alignment as guided by "Resource Rational Contractualism" (RRC).
But wait -- what's that? A 🧵.
🚨New paper! We know models learn distinct in-context learning strategies, but *why*? Why generalize instead of memorize to lower loss? And why is generalization transient?
Our work explains this & *predicts Transformer behavior throughout training* without its weights! 🧵
1/
Excited to share a new CogSci paper co-led with @benpry.bsky.social!
Once a cornerstone for studying human reasoning, the think-aloud method declined in popularity as manual coding limited its scale. We introduce a method to automate analysis of verbal reports and scale think-aloud studies. (1/8)🧵
🤔🤖Most AI systems assume there’s just one right answer—but many tasks have reasonable disagreement. How can we better model human variation? 🌍✨
We propose modeling at the individual-level using open-ended, textual value profiles! 🗣️📝
arxiv.org/abs/2503.15484
1/13 New Paper!! We try to understand why some LMs self-improve their reasoning while others hit a wall. The key? Cognitive behaviors! Read our paper on how the right cognitive behaviors can make all the difference in a model's ability to improve with RL! 🧵
There are also many papers on information diffusion…..
Noah way!
So where’s your pack of AI researchers that defy categorization?
But @roydanroy.bsky.social where do I belong?? A crisis of affective self theory.
@tomerullman.bsky.social I made one for your lab!
This seems like a good first post: what happens to psychology in the LLM future? I ponder… open.substack.com/pub/noahgood...
