ASTRA USES A NEW REASONING TECHNIQUE CALLED "RECURRENT DEPTH". TL;DR: > instead of stacking more Transformer layers, it loops the model's hidden state through the same layer multiple times. some of the reasoning happens internally rather than writing it out as chain-of-thought tokens. why this may help: > more internal reasoning compute without generating a huge CoT > it doesn't keep adding tokens to the KV cache > as @eliebakouch pointed out in his post, recurrent depth isn't necessarily…

Stephanie Palazzolo@steph_palazzoloSep 2

OpenAI’s Astra AI uses a new reasoning approach called “recurrent depth.” Though it can help model costs and performance, researchers are concerned bc it obscures a model’s thinking process, making it more difficult to monitor. w/ @amir @rocketalignment https://www.theinformation.com/articles/secret-technique-behind-openais-astra-model-sparks-security-concerns