Logo Logo
Hilfe
Hilfe
Switch Language to English

Stenzel, Gerhard; Zorn, Maximilian ORCID logoORCID: https://orcid.org/0009-0006-2750-7495; Altmann, Philipp ORCID logoORCID: https://orcid.org/0000-0003-1134-176X; Mansky, Maximilian; Kölle, Michael und Gabor, Thomas ORCID logoORCID: https://orcid.org/0000-0003-2048-8667 (2024): Self-Replicating Prompts for Large Language Models: Towards Artificial Culture. ALife 2024: The International Conference for Artificial Life, Kopenhagen, Dänemark, 22. - 26. Juli 2024. Cambridge: MIT. [PDF, 515kB]

[thumbnail of isal_a_00813.pdf]
Vorschau
Creative Commons: Namensnennung 4.0 (CC-BY)
Veröffentlichte Version

Abstract

We consider various setups where large language models (LLMs) communicate solely with themselves or other LLMs. In accordance with similar results known for program representations (like λ-expressions or automata), we observe a natural tendency for the evolution of self-replicating text pieces, i.e., LLM prompts that cause any receiving LLM to produce a response similar to the original prompt. We argue that the study of these self-replicating patterns, which exist in natural language and across different types of LLMs, may have important implications on artificial intelligence, cultural studies, and related fields.

Dokument bearbeiten Dokument bearbeiten