3 ms·
Have it emit plausible-looking x64 instructions by training on lots of executables, get a program out which has some behaviour. Might be worth seed funding at t
by JonChesterfield 2y ago
Have it emit plausible-looking x64 instructions by training on lots of executables, get a program out which has some behaviour. Might be worth seed funding at the moment.
- pjmlp 2y agoYeah, this is however the same kind of discussion as back in the day Assembly developers not trusting FORTRAN compilers, so it is a matter of time, and funding.
- JonChesterfield 2y agoThe Fortran compilers were trying to get the answer right whereas the proposed funding void would at best be trying to avoid a segv. What probably does have real merit is tying a superoptimiser to a LLM, provided you've got the SAT solver included in the mix as well to know if it worked.