maš¯•¸pool on Nostr: Mixing formal theorem proofing with deep learning seems promising. They complement ...
Mixing formal theorem proofing with deep learning seems promising. They complement each other. Theorem proving "keeps it real" while deep learning provides heuristics.
DeepSeek-Prover: Advancing Theorem Proving in LLMs through Large-Scale Synthetic Data
https://arxiv.org/abs/2405.14333#Lean4 #LLM #TheoremProving #DeepLearning #DL
Published at
2024-10-14 16:33:10Event JSON
{
"id": "522d543d69cac9340232b2e28c7b4c727f10351cdc6ff8525a4b8d1d14803313",
"pubkey": "9bea6901445ff93697e6e05a794147d7962522fb5725d6e5c55130a8ddbf911a",
"created_at": 1728923590,
"kind": 1,
"tags": [
[
"t",
"dl"
],
[
"t",
"deeplearning"
],
[
"t",
"TheoremProving"
],
[
"t",
"llm"
],
[
"t",
"Lean4"
],
[
"proxy",
"https://mathstodon.xyz/users/maxpool/statuses/113306736428226834",
"activitypub"
]
],
"content": "Mixing formal theorem proofing with deep learning seems promising. They complement each other. Theorem proving \"keeps it real\" while deep learning provides heuristics. \n\nDeepSeek-Prover: Advancing Theorem Proving in LLMs through Large-Scale Synthetic Data\nhttps://arxiv.org/abs/2405.14333\n\n#Lean4 #LLM #TheoremProving #DeepLearning #DL",
"sig": "cd18cb100f8b578f07056366d7dd3c8bacda8b7d0e75e8785ce8193008d4d94ba8a035427a71388914228b382f2d6225830349b54262016a40da407e25164940"
}
