Grounding Large Language Models in Interactive Environments with Online Reinforcement Learning - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2023

Grounding Large Language Models in Interactive Environments with Online Reinforcement Learning

Abstract

Recent works successfully leveraged Large Language Models' (LLM) abilities to capture abstract knowledge about world's physics to solve decision-making problems. Yet, the alignment between LLMs' knowledge and the environment can be wrong and limit functional competence due to lack of grounding. In this paper, we study an approach (named GLAM) to achieve this alignment through functional grounding: we consider an agent using an LLM as a policy that is progressively updated as the agent interacts with the environment, leveraging online Reinforcement Learning to improve its performance to solve goals. Using an interactive textual environment designed to study higher-level forms of functional grounding, and a set of spatial and navigation tasks, we study several scientific questions: 1) Can LLMs boost sample efficiency for online learning of various RL tasks? 2) How can it boost different forms of generalization? 3) What is the impact of online learning? We study these questions by functionally grounding several variants (size, architecture) of FLAN-T5.
Fichier principal
Vignette du fichier
extended_version.pdf (3.09 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Licence : CC BY - Attribution

Dates and versions

hal-03970122 , version 1 (02-02-2023)
hal-03970122 , version 2 (25-05-2023)
hal-03970122 , version 3 (06-09-2023)

Licence

Attribution

Identifiers

Cite

Thomas Carta, Clément Romac, Thomas Wolf, Sylvain Lamprier, Olivier Sigaud, et al.. Grounding Large Language Models in Interactive Environments with Online Reinforcement Learning. International Conference on Machine Learning 2023, Jul 2023, Honololu, Hawaii, United States. ⟨hal-03970122v3⟩
130 View
158 Download

Altmetric

Share

Gmail Facebook X LinkedIn More