higuidebot

What do you think a "better" result would be here? Better by what metric?

Definitely plausible but it underrates the changes in actual content. It's not just SEO and titling, it's actual videos that have "stepsis" etc. as themes.

I am the author, it's just those 3 terms for the tSNE cluster. Sorry, I can tell from some of the comments here the graphs need to be clearer. "Stepsis" is indicative of incest IMO, the "step" is a fig leaf.

It's not quite "random" to look at the most intentionally directed legislation of several years. What's your explanation? Genuinely curious. One way or another the clusters do exist, the trends do exist in both content and titling convention

SEO contributes for sure, but I would reverse the statement here - it's more about new content and less about SEO. There's a feedback loop race to the bottom dynamic regardless

>For their category "Sexual Violence", they include both violent terms ("woman being raped", "torture porn"), and non-violent terms ("stepsis")

False, the final term is "incest", not "stepsis". All of the above are semantically valid for "sexual violence".

Lol @ R1's description of my Github ... thanks Deepseek, very cool!

No need to do anything in particular! Perhaps interesting to observe

Ha, if steps have consistent distances you could take the average distance at step X and generate a step of that length in some direction and be ~approximately correct regardless of the actual value

Very cool!

Well, you are certainly correct about how cosine sim would apply to the text embeddings, but I disagree about how useful that application is to our understanding of the model.

> In this case, cosine distance one would be in a case when it repeats word-by-word. It is not even a "similar thought" but some sort of LLM's OCD.

Observing that would be helpful in our understanding of the model!

> For anything else... cosine similarity says little. Sometimes, two steps can have opposite consultation, but they have very high cosine similarity. In another case, it can just expand on the same solution but use different vocabulary or look from another angle.

Yes, that would be good to observe also! But here I think you undervalue the specificity of the OAI embeddings model, which has 3072 dimensions. That's quite a lot of information being captured.

> A more robust approach would be to give the whole reasoning to an LLM and ask to grade according to a given criterion (e.g. "grade insight in each step, from 1 to 5").

Totally disagree here, using embeddings is much more reliable / robust, I wouldn't put much stock in LLM output, too much going on