do large-scale vision models represent the 3D structure of objects? excited to share our benchmark: multiview object consistency in humans and image models (MOCHI) with Stephanie Fu Yutong Bai Thomas O'Connell Yoni Friedman Nancy Kanwisher @[email protected] Josh Tenenbaum and Alexei Efros 1/👀

Really excited to be a part of this amazing lineup!

📢New paper: Many companies and papers have claimed AI can automate science. How can we evaluate these claims? Today, we introduce CORE-Bench: a benchmark to measure if AI agents can automate reproducing a paper given access to its code and data. arxiv.org/pdf/2409.11363…

Great paper, and so timely!

New results about CoT! For the cogsci crowd: I think a lot to think about re why Chain of Thought is more useful to math/logic than other domains. For the perhaps niche-but-enthusiastic meta-analysis crowd: I hope we can be part of a trend towards more meta-analyses in NLP!

Exciting news🎉 We're now accepting applications for CBS-NTT postdoc fellows in the Physics of Intelligence at Harvard (rb.gy/gs7z3r) Join our vibrant interdisciplinary community at Harvard to shape this exciting emerging field together! Link below:

We are currently hiring for an open research scientist position working on my team in multi-agent artificial general intelligence: boards.greenhouse.io/deepmind/jobs/…

This massive analysis points to something fundamental about what really affects the results of studies - Effect of *population* (culture, age, etc.) is relatively small - Effect of design (e.g., the specifics of the stimuli) is considerably larger pnas.org/doi/10.1073/pn…