3 ms·
I don’t understand how this works for anything but prose. Is that the point? In any code or structured output, there just isn’t the flexibility, and depending o
by andy99 1mo ago
I don’t understand how this works for anything but prose. Is that the point? In any code or structured output, there just isn’t the flexibility, and depending on how the user requests the output be constrained there is even less (“answer only True or False”). So is it just chat responses? If I ask the API to tell me a story about Alice and Bob then it watermarks it, but when I ask it some implausibly constrained thing like write a story about Alice and Bob with each word starting in rotation with the letters alicebob, does it try to do so and hope there are roughly équiprobable tokens regularly?
- chrisjj 1mo ago> In any code or structured output, there just isn’t the flexibility Variable name perversion incoming...
- smallerize 1mo agoYes, and it says that in https://www.anthropic.com/news/claude-text-watermark https://www.anthropic.com/news/claude-text-watermark
- andy99 1mo agoI should have read that, it’s actually quite reasonable and I don’t really understand the objections in TFA having read it. > One of my fundamental problem with this is that no two synonyms carry the exact same meaning. “He leaped at the chance” and “He jumped at the opportunity” are very similar sentences expressing the same general sentiment, but they are not the same. The exact words we choose when writing matter. Doesn’t make sense at all in light of the actual approach, they’re just choosing a different RNG. It’s not like they’re corrupting it by flipping words. Should add I don’t support the watermarking and requiring it is idiotic.
- krackers 1mo agoI don't understand Gruber's points either, I wonder if there is some fundamental technical misunderstanding. Does he think that the logits should be sampled from in a "pure" manner without introducing any other bias? Does he know that there's already a sampling temperature, and that most providers have probably moved on to sampling strategies other than top-k? Does he know that the word choices have already been altered irreversibly during RLHF which is how you get the obvious Claudism like "load bearing" and "seams"? Perhaps it would be useful to publish examples of samples with/without watermark. I'd suspect that the variability from simply sampling repeated times would dwarf any semantic differences you'd detect with the watermark.
- troupo 1mo ago> I don't understand Gruber's points either, Gruber's point is to bash the EU. He couldn't care much about anything else.
- smallerize 1mo agoI think Anthropic should have put all the info into one blog post. Splitting it up is really confusing people.
- Cakez0r 1mo agoThis could be why claude code has recently started to write reams of inane comments alongside the code it generates.