Sharon Machlis on Nostr: “What tasks can an #LLM do well?” turns out to be a complicated question. This ...
Published at
2024-11-13 12:20:48Event JSON
{
"id": "b6715982ae3558c0befbdcd41678f2abb107e0b1fb115f6254ffe2505d4db9fc",
"pubkey": "1c26d7da08016e96038cde0d08b3670b23ea4d23e798029d0e74f65832f80be1",
"created_at": 1731500448,
"kind": 1,
"tags": [
[
"t",
"llm"
],
[
"t",
"llms"
],
[
"t",
"genai"
],
[
"proxy",
"https://masto.machlis.com/users/smach/statuses/113475613366958322",
"activitypub"
]
],
"content": "“What tasks can an #LLM do well?” turns out to be a complicated question. This article looks at whether “emergence” - when #LLMs seem to suddenly develop new skills - is truly a thing:\nhttps://cacm.acm.org/news/is-it-possible-to-truly-understand-performance-in-llms/\n#GenAI",
"sig": "0583c8194bd3a481a3becc657afeda6b0fc75cc74435cc138c47b71fe1bcf4e5b2595215b623c362647e159d4762581723be725490c28fd6c5ba30f0a2effdb7"
}
