Hacker News
Favorites
Setup
☰
Home
Favorites
Setup
Comment by wwind123 |
original
|
Senior SWE-Bench: open-source benchmark that assesses agents as senior engineers
[−]
wwind123
· 2026-07-02 Thu 06:37 UTC ·
link
fave
This kind of approach would generally still need human guidance, otherwise these models might get stuck in weird niche corners of the problem space that would not be relevant to any real world project.
[−]
ben_w
· 2026-07-02 Thu 07:19 UTC ·
link
fave
We could call this "reinforcement learning from human feedback" (RLHF) :)
https://en.wikipedia.org/wiki/Reinforcement_learning_from_hu...
https://en.wikipedia.org/wiki/Reinforcement_learning_from_hu...