Language Agents as Optimizable Graphs
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910572702859264 |
|---|---|
| author | Zhuge, Mingchen Wang, Wenyi Kirsch, Louis Faccio, Francesco Khizbullin, Dmitrii Schmidhuber, Jürgen |
| author_facet | Zhuge, Mingchen Wang, Wenyi Kirsch, Louis Faccio, Francesco Khizbullin, Dmitrii Schmidhuber, Jürgen |
| contents | Various human-designed prompt engineering techniques have been proposed to improve problem solvers based on Large Language Models (LLMs), yielding many disparate code bases. We unify these approaches by describing LLM-based agents as computational graphs. The nodes implement functions to process multimodal data or query LLMs, and the edges describe the information flow between operations. Graphs can be recursively combined into larger composite graphs representing hierarchies of inter-agent collaboration (where edges connect operations of different agents). Our novel automatic graph optimizers (1) refine node-level LLM prompts (node optimization) and (2) improve agent orchestration by changing graph connectivity (edge optimization). Experiments demonstrate that our framework can be used to efficiently develop, integrate, and automatically improve various LLM agents. The code can be found at https://github.com/metauto-ai/gptswarm. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_16823 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Language Agents as Optimizable Graphs Zhuge, Mingchen Wang, Wenyi Kirsch, Louis Faccio, Francesco Khizbullin, Dmitrii Schmidhuber, Jürgen Artificial Intelligence Computation and Language Machine Learning Multiagent Systems Various human-designed prompt engineering techniques have been proposed to improve problem solvers based on Large Language Models (LLMs), yielding many disparate code bases. We unify these approaches by describing LLM-based agents as computational graphs. The nodes implement functions to process multimodal data or query LLMs, and the edges describe the information flow between operations. Graphs can be recursively combined into larger composite graphs representing hierarchies of inter-agent collaboration (where edges connect operations of different agents). Our novel automatic graph optimizers (1) refine node-level LLM prompts (node optimization) and (2) improve agent orchestration by changing graph connectivity (edge optimization). Experiments demonstrate that our framework can be used to efficiently develop, integrate, and automatically improve various LLM agents. The code can be found at https://github.com/metauto-ai/gptswarm. |
| title | Language Agents as Optimizable Graphs |
| topic | Artificial Intelligence Computation and Language Machine Learning Multiagent Systems |
| url | https://arxiv.org/abs/2402.16823 |