News for hackers. Stuff that matters.

A comment on OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computers. Parent comment

Re:OSWorld: Benchmarking Multimodal Agents for Open-Ended Ta... by ec109685 on Sunday April 28, 2024 @07:15PM (#40192850)
What’s the most complicated thing it was able to debug?
[ Parent | on HN ]
Re:OSWorld: Benchmarking Multimodal Agents for Open-Ended Ta... by stavros on Sunday April 28, 2024 @07:17PM (#40192875)
I had a library missing and it started using ldd to see where things were linked, and it changed the library path to the proper location, it was impressive.
[ Parent | on HN ]
Re:OSWorld: Benchmarking Multimodal Agents for Open-Ended Ta... by bamboozled on Sunday April 28, 2024 @08:20PM (#40193183)
Sounds like the state of your computer is the real problem here.
[ Parent | on HN ]
Re:OSWorld: Benchmarking Multimodal Agents for Open-Ended Ta... by stavros on Monday April 29, 2024 @03:28AM (#40195507)
As opposed to?
[ Parent | on HN ]
Re:OSWorld: Benchmarking Multimodal Agents for Open-Ended Ta... by bamboozled on Monday April 29, 2024 @05:37AM (#40196331)
https://nixos.org/

But just in general, I've been a Linux pro for about 15 years, it's extremely rare I'd have to manually link anything?

[ Parent | on HN ]
Re:OSWorld: Benchmarking Multimodal Agents for Open-Ended Ta... by stavros on Monday April 29, 2024 @11:28AM (#40199589)
Well, when you need to run software that needs a library and your system doesn't have it because the library is too old, what do you do?
[ Parent | on HN ]