In authors or contributors

2 resources

  • Noah Shinn, Federico Cassano, Beck Labas...
    |
    May 21st, 2023
    |
    preprint
    Noah Shinn, Federico Cassano, Beck Labas...
    May 21st, 2023

    Large language models (LLMs) have been increasingly used to interact with external environments (e.g., games, compilers, APIs) as goal-driven agents. However, it remains challenging for these language agents to quickly and efficiently learn from trial-and-error as traditional reinforcement learning methods require extensive training samples and expensive model fine-tuning. We propose Reflexion, a novel framework to reinforce language agents not by updating weights, but instead through...

  • Noah Shinn, Federico Cassano, Beck Labas...
    |
    May 21st, 2023
    |
    preprint
    Noah Shinn, Federico Cassano, Beck Labas...
    May 21st, 2023

    Large language models (LLMs) have been increasingly used to interact with external environments (e.g., games, compilers, APIs) as goal-driven agents. However, it remains challenging for these language agents to quickly and efficiently learn from trial-and-error as traditional reinforcement learning methods require extensive training samples and expensive model fine-tuning. We propose Reflexion, a novel framework to reinforce language agents not by updating weights, but instead through...

Last update from database: 28/12/2024, 23:15 (UTC)
Powered by Zotero and Kerko.