Re:GPT-5.2
by fuddle on Thursday December 11, 2025 @02:49PM
(#46236242)
I don't think SWE Verified is an ideal benchmark, as the solutions are in the training dataset.
[ Parent
| on HN ]
Re:GPT-5.2
by joshuahedlund on Thursday December 11, 2025 @03:33PM
(#46236752)
I would love for SWE Verified to put out a set of fresh but comparable problems and see how the top performing models do, to test against overfitting.