Comment by magicalhippo
1 year ago
That's just because we twisted it's arm. One could for example feed the reversed input after, ie abc|cba where | is a special token. That would allow it to react to any part of the message.
1 year ago
That's just because we twisted it's arm. One could for example feed the reversed input after, ie abc|cba where | is a special token. That would allow it to react to any part of the message.
I think this might be key, in addition to some landmark tokens to quickly backtrack to. The big question is how to train such model.
There is a recent paper from Meta that propose a way to train a model to backtrack its generation to improve generation alignment [0].
[0] https://arxiv.org/html/2409.14586v1