binyu

This is particularly relevant because the guy used Andrej Karphaty's "autoresearch" [1] approach, which is extremely powerful when applied to problems that generalize well to it. [1] h…

Can you provide an example or a reproducible way to trigger it?

That behavior was only observed with the short lived Fable preview, Opus and other models do not seem to pose such limitation.

> This would be a great topic but this specific article isn't worth your time. Why? Do you think it reports inaccurate facts? Personally, I found it interesting and easy to read.

Personally, I've tried to squeeze more tok/s for a single DGX Spark deployment and DeepSeek V4 Flash but only got marginal improvements. There's work to do on fusing kernels and other o…

> Now at 40-50tok/s generation and ~2000 tok/s Not clear how you went from ~11-14 to ~40-50 tok/s. Is it by running the quant native model and adding a second Spark? Cheers

> Fable kept trying to propose fixes and I couldn't get it to construct e2e chain What approach did you start with? Can you elaborate?

Depends of what the proof of concept is about. It could be just a toy example, e.g. a RCE that opens the calculator app or something much more nefarious, like returning a root shell and would still fa…

I was able to use Fable to generate PoC for several classes of vulnerabilities and I didn't observe the model refusing to engage in detailed analysis to come up with creative approaches, the very…

> We reviewed a demonstration of this specific technique being used to identify a small number of previously known, minor vulnerabilities. "small number", "previously known"+&qu…

> This is ignoring the fact that the main reason retired phones are e-waste is proprietary firmware blobs and locked-down systems Couldn't Google somehow fix this? Since they control the subst…