The Unreasonable Effectiveness of Eccentric Automatic Prompts
Large Language Models (LLMs) have demonstrated remarkable problem-solving and basic mathematics abilities. However, their efficacy is highly contingent on the formulation of the prompt. This study endeavors to quantify the influence of incorporating “positive thinking” into the system message of the prompt, then compare that to systematic pro ...
https://arxiv.org/html/2402.10949v2
brunoamaral /
npub1ykn…5r6f
2024-03-04 10:10:00
Author Public Key
npub1yknpd722us9r25kqx7mvendmfve3uc2yhkr9k8p76f04svluzucqfp5r6fPublished at
2024-03-04 10:10:00Event JSON
{
"id": "9c321ff1158bbe13d469624ce1cfb873670c4deacfd516ce9475a33b8b6f0228",
"pubkey": "25a616f94ae40a3552c037b6cccdbb4b331e6144bd865b1c3ed25f5833fc1730",
"created_at": 1709543400,
"kind": 1,
"tags": [
[
"proxy",
"https://mastodon.social/users/brunoamaral/statuses/112036636277783138",
"activitypub"
]
],
"content": "The Unreasonable Effectiveness of Eccentric Automatic Prompts\n\nLarge Language Models (LLMs) have demonstrated remarkable problem-solving and basic mathematics abilities. However, their efficacy is highly contingent on the formulation of the prompt. This study endeavors to quantify the influence of incorporating “positive thinking” into the system message of the prompt, then compare that to systematic pro ...\nhttps://arxiv.org/html/2402.10949v2",
"sig": "4371416fe57347781d2450966337b991e22cc36aa2a4032a8c93bda560d0aba9e75606bbcf454595d49a0367e41387f1efd8796fb282c19fdcff60b64f4ec5f7"
}