Inertia Cottontail on Nostr: OpenAI is now resorting to cheating on benchmarks to make the numbers look better. ...
OpenAI is now resorting to cheating on benchmarks to make the numbers look better. Lol
Published at
2025-01-20 15:41:50Event JSON
{
"id": "bd3cb90dfee6807ba02c65b4001abbb7e33ceaf86c8a782f5ff23e7f569e6cd4",
"pubkey": "e18127cdb6bc491133c63a14fc421a9291029cafa2245f70a5769febbfae2d70",
"created_at": 1737387710,
"kind": 1,
"tags": [
[
"imeta",
"url https://treebrary.pone.social/media_attachments/files/113/861/438/197/138/129/original/d1d47cf611ecc6d3.png",
"m image/png",
"dim 868x123",
"blurhash U9QvwRIUt7RjIUofxuof~qofRjof-;ofRjof"
],
[
"proxy",
"https://pone.social/users/inertia/statuses/113861440986684538",
"activitypub"
]
],
"content": "OpenAI is now resorting to cheating on benchmarks to make the numbers look better. Lol\n\nhttps://treebrary.pone.social/media_attachments/files/113/861/438/197/138/129/original/d1d47cf611ecc6d3.png",
"sig": "92375538d5847f658637ac10d388efb3991bd0c5665ed250139d1f5393efedbc233ba0d2327407b44e13528ce747a43297ed20adc5f7d38df4409cc3b553ca49"
}