jb55 on Nostr: frontier labs are cookin reinforcement learning with verifiable feedback, I can feel ...
frontier labs are cookin reinforcement learning with verifiable feedback, I can feel it. LLMs + superhuman reasoning with RL is ggs.
Published at
2025-03-20 16:01:33Event JSON
{
"id": "d5af1a8dbd6eb63ccdba149ce6c89e0b744c0746cdf36f526f0982894d39cc76",
"pubkey": "32e1827635450ebb3c5a7d12c1f8e7b2b514439ac10a67eef3d9fd9c5c68e245",
"created_at": 1742486493,
"kind": 1,
"tags": [
[
"client",
"Damus Notedeck"
]
],
"content": "frontier labs are cookin reinforcement learning with verifiable feedback, I can feel it. LLMs + superhuman reasoning with RL is ggs.",
"sig": "2af8264a7124339df9b0509b21f2e5ebdf6fbca2682800006115e1d0e7edbb1288605bf4e58a7a4aacdd1e0e1524d76539a8be1e7e172abb384045235d14724a"
}