<p><pre><code> > ...our results give: ... (3) a provable advantage of chain-of-thought, exhibiting a task that becomes exponentially easier with chain-of-thought.
</code></pre>
It would be good to also prove that there is no task that becomes exponentially harder with chain-of-thought.
Loosely related thought: A year ago, there was a lot of talk about the Mamba SSM architecture replacing transformers. Apparently that didn't happen so far.
Quanta magazine has an article that explains in plain words what the researchers were trying to do : <a href="https://www.quantamagazine.org/chatbot-software-begins-to-face-fundamental-limitations-20250131/" rel="nofollow">https://www.quantamagazine.org/chatbot-software-begins-to-fa...</a>
Huh. I just skimmed this and quickly concluded that it's definitely <i>not</i> light reading.<p>It sure looks and smells like good work, so I've added it to my reading list.<p>Nowadays I feel like my reading list is growing faster than I can go through it.