Comment on Semantic ablation: Why AI writing is boring and dangerous

lvxferre@mander.xyz ⁨9⁩ ⁨hours⁩ ago

I believe that good communication has four attributes.

  1. It’s approachable: it demands from the reader (or hearer, or viewer) the least amount of reasoning and previous knowledge, in order to receive the message.
  2. It’s succinct: it demands from the reader the least amount of time.
  3. It’s accurate: it neither states nor implies (for a reasonable = non-assumptive receiver) anything false.
  4. It’s complete: it provides all relevant information concerning what’s being communicated.

However no communication is perfect and those four attributes are in odds with each other: if you try to optimise your message for one or more of them, the others are bound to suffer.

Why this matters here: it shows the problem of ablation is unsolvable. Even if generative models were perfectly competent at rephrasing text (they aren’t), simply by asking them to make the text more approachable, you’re bound to lose info or accuracy.

I’d also argue “semantic ablation” is a way better concept than “hallucination”. The later is not quite “additive error”; it’s a misleading metaphor for output that is generated by the model the same way as the rest, but it happens to be incorrect when interpreted by human beings.

source
Sort:hotnewtop