First screen, captured .
758 points by beefman | 64 comments
Project page for Visual Anagrams, a University of Michigan research paper using pretrained diffusion models to generate multi-view optical illusions.
684 points by jackcook | 148 comments
Project page for MeshGPT, a research paper from TU Munich, Politecnico di Torino and Audi that generates triangle meshes using decoder-only transformers.
85 points by merksittich | 64 comments
Epoch AI report by David Owen presenting InnovationEval, finding early evidence that AI models cannot yet automate AI R&D or match human algorithmic innovation.
334 points by Cynddl | 333 comments
arXiv paper finding that fine-tuning language models to be warm and empathetic raises error rates and makes them more sycophantic on safety-critical tasks.
14 points by fagnerbrack | 7 comments
A blog post by Fagner Brack on Medium describing what an AI-native approach to software development with stable infrastructure looks like.
549 points by chmaynard | 134 comments
Rachel Thomas blog post on how a high-profile Nature paper using a Transformer to predict enzyme functions contained many errors that received far less attention than the original result.
453 points by bx376 | 663 comments
An Economist Free exchange column arguing that artificial intelligence is losing hype and asking whether that signals the technology will eventually succeed.
684 points by jackcook | 148 comments
Project page for MeshGPT, a research paper from TU Munich, Politecnico di Torino and Audi that generates triangle meshes using decoder-only transformers.