Controllable-CoT leads to covert reasoning capabilities

SummaryI measure GPT-6 Astra’s performance on multi-hop tasks when prompted with a secondary CoT-control instruction: to reason using only dots, or to reason steganographically. Astra demonstrates covert reasoning capabilities with task performance beating that when using no reasoning or filler tokens for reasoning.This work agrees with findings from Astra is much better at reasoning with filler tokens than previous models but has the model generate its own reasoning and provide it as part of th...