Live data from Hacker News

Cross-Model Void Convergence: GPT-5.2 and Claude Opus 4.6 Deterministic Silence

zenodo.org

31–34 of 34 posts

Re: Cross-Model Void Convergence: GPT-5.2 and Claude Opus 4.6 Deterministic Silence

#31
I like that of the 6 submissions that this ‘researcher’ has submitted, 5 of them are about how some language models will output what looks like nothing if you tell them to output nothing. I’m assuming this sort of phenomenon is why places like arXiv are seeing their costs explode.

https://zenodo.org/search?q=metadata.creators.person_or_org....

Re: Cross-Model Void Convergence: GPT-5.2 and Claude Opus 4.6 Deterministic Silence

#32
post #30
post #28

Earlier quoted context omitted.

Out of curiosity, are there any sources to there being a significant amount of other steps before being fed into the weights Security guards / ... are the obvious ones, but do you mean they have branching early on to shortcut certain prompts?

> do you mean they have branching early on to shortcut certain prompts? Putting a classifier in front of a fleet of different models is a great way to provide higher quality results and spend less energy. Classification is significantly cheaper than generation and it is the very first thing you would do here. A default, catch-all model is very expensive, but handles most queries reasonably well. The game from that po…

Aren't you describing why they use mixture of experts? Where a sub-set of weights are activated depending on the query?

Re: Cross-Model Void Convergence: GPT-5.2 and Claude Opus 4.6 Deterministic Silence

#33
post #29

This is 100% crank “science” that has wrapped up a banal finding in big words and LaTeX. The claim is roughly as exciting as, “some computer programs print nothing to stdout.” The output shown is not “null” or “void”. It is the empty string, which these LLMs are perfectly capable of outputting. Technically, it outputs the stop token, analogous to \0 at the end of a C string.

It looks like jargon and LaTeX are the future of spam given how readily pepper treat anything framed in them as legitimate.

Can’t wait to start getting spam emails about exogenous upregulation of androgen receptor density within the penile vasculature to potentiate trophic tissue remodeling from penus-enlargement.biz

Re: Cross-Model Void Convergence: GPT-5.2 and Claude Opus 4.6 Deterministic Silence

#34
post #29

This is 100% crank “science” that has wrapped up a banal finding in big words and LaTeX. The claim is roughly as exciting as, “some computer programs print nothing to stdout.” The output shown is not “null” or “void”. It is the empty string, which these LLMs are perfectly capable of outputting. Technically, it outputs the stop token, analogous to \0 at the end of a C string.

Finally someone called it out, it's crazy that such posts get to the HN front page. Zenodo gets so much AI slop spam nowadays.
Post reply on HN