it’s straightforward, just detect if perplexity is too high. can they detect perplexity too high? like… probably, they’re prediction models, but it’s not clear to me whether or to what degree they notice previous prediction errors when predicting additional tokens
it’s straightforward, just detect if perplexity is too high. can they detect perplexity too high? like… probably, they’re prediction models, but it’s not clear to me whether or to what degree they notice previous prediction errors when predicting additional tokens