این رفیقمون جاسپر لو اومده دلیل واقعی کارکردن «زنجیره تفکر» یا همون CoT رو تو مدل‌های هوش مصنوعی با یه تحلیل مشتی باز کرده

It took me a long time to build an intuition for why CoT works. My thinking was always.. if the model can predict it downstream of 10k thinking tokens, it should have been able to predict it from the outset too. My intuition now is: - During inference, the correct paths are indeed somewhere in the hidden states, represented purely as probabilities - However, in the process of sampling, we're forced to materialize just one path. This is destructive -- a 30% chance of ending up at the answer can become 0 if we sample the wrong token. - The constant backtracking reasoning models do protect against this. Every "wait" or "but" is another chance for a shot on target. - By the time models exhaust their reasoning budget, they've already seen a bunch of possible answers - And since these models are also generally better at verifying answers than generating them, the chances of choosing the correct path, conditioned on this prefix, are much higher than it was at the start.