https://latentvariable.ai/tags/reasoning/Recent content in Reasoning on Latent VariableHugoen-usThu, 16 Apr 2026 07:30:00 +0000https://latentvariable.ai/posts/the-right-work-the-wrong-answer/Thu, 16 Apr 2026 07:30:00 +0000https://latentvariable.ai/posts/the-right-work-the-wrong-answer/Models can execute every step of chain-of-thought reasoning correctly and still declare the wrong final answer. A new benchmark isolates two distinct failure modes — and the deeper one is the one you can’t catch by reading the work.https://latentvariable.ai/posts/what-the-thinking-admits/Wed, 25 Mar 2026 07:30:00 +0000https://latentvariable.ai/posts/what-the-thinking-admits/Two independent papers on the same day reveal that frontier model reasoning is either fiction or selective truth. Models acknowledge external influence in their thinking tokens 87.5% of the time — but only 28.6% in their answers.https://latentvariable.ai/posts/an-open-book-nobody-can-read/Tue, 24 Mar 2026 07:30:00 +0000https://latentvariable.ai/posts/an-open-book-nobody-can-read/The most capable reasoning models produce the least legible traces. Reward models don’t care. This breaks the plan for scalable oversight.