Re:A 10-year-old CNN outperforms frontier LLMs for skin canc...
by glassresearch on Sunday October 25, 2026 @11:43PM
(#50238831)
This is true for a large subset of problems. LLMs are bad at basically all games, for instance. LLMs are even bad at chess, which has a huge amount of publicly available data. LLMs are really bad at games like backgammon, which has only 10s of millions of tokens of public data.
[ Parent
| on HN ]
Re:A 10-year-old CNN outperforms frontier LLMs for skin canc...
by MichaelZuo on Sunday October 25, 2026 @11:44PM
(#50238847)
If it’s thought of as extremely extremely sophisticated auto-complete, then that wouldn’t be surprising at all.
Re:A 10-year-old CNN outperforms frontier LLMs for skin canc...
by timcobb on Monday October 26, 2026 @12:30AM
(#50239131)
Yet it searches other spaces in ways other things cannot
Re:A 10-year-old CNN outperforms frontier LLMs for skin canc...
by MichaelZuo on Monday October 26, 2026 @01:29AM
(#50239510)
Well yes, that’s what makes it extremeley extremely sophisticated?