mlsanityData to 14 Oct 2026

LaGO: Latent Action Guidance for Online Reinforcement Learning

arXiv 2606.24669 ↗First posted 23 Jun 2026cs.AIcited by 0

On the map: Deep reinforcement learning methods → RL training of language and agent models

Abstract (arXiv)

Large language models (LLMs) have shown strong potential for planning and sequential decision-making, but prior work often relies on using them as direct controllers, which requires precise action generation and can be unreliable in practice. This paper proposes Latent Action Guidance for Online Reinforcement Learning (LaGO), a framework that uses a pretrained LLM as a latent action prior to softly guide online policy optimization, rather than treating the LLM as an explicit planner or controller. Experiments on both a discrete-control benchmark, CLEVR-Robot, and a continuous-control benchmark, Meta-World, demonstrate that LaGO consistently improves both reward and success rate over Vanilla PPO. In particular, LaGO increases the average success rate from 15.1% to 27.2% on CLEVR-Robot and from 2.7% to 15.2% on Meta-World. Our analysis further shows that stronger pretrained LLMs provide more effective guidance, suggesting that LLM knowledge can improve planning and online decision-making.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL training of language and agent models, one of 15 directions in Deep reinforcement learning methods. Builds on 9 of its 14 references that are in the map. First posted on arXiv 23 Jun 2026.

Builds on

9 of 14 references.

References · 9

  1. 2023Voyager: An Open-Ended Embodied Agent with Large Language Modelscited by 2,191
  2. 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,204
  3. 2023Reasoning with Language Model is Planning with World Modelcited by 861
  4. 2022Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Taskcited by 431
  5. 2023Grounding Large Language Models in Interactive Environments with Online Reinforcement Learningcited by 195
  6. 2025WebEvolver: Enhancing Web Agent Self-Improvement with Coevolving World Modelcited by 61
  7. 2023Emergent Representations of Program Semantics in Language Models Trained on Programscited by 30
  8. 2026OccuBench: Evaluating AI Agents on Real-World Professional Tasks via Language Environment Simulationcited by 7
  9. 2024Knowledgeable Agents by Offline Reinforcement Learning from Large Language Model Rolloutscited by 6

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026A Subgoal-driven Framework for Improving Long-Horizon LLM Agents
  2. 2024Survey on Large Language Model-Enhanced Reinforcement Learning: Concept, Taxonomy, and Methods
  3. 2025LLM-Guided Reinforcement Learning: Addressing Training Bottlenecks through Policy Modulation
  4. 2025LANPO: Bootstrapping Language and Numerical Feedback for Reinforcement Learning in LLMs
  5. 2025LLM-Guided Task- and Affordance-Level Exploration in Reinforcement Learning
  6. 2026PriorZero: Bridging Language Priors and World Models for Decision Making
  7. 2025Response-Level Rewards Are All You Need for Online Reinforcement Learning in LLMs: A Mathematical Perspective
  8. 2024Plan-Seq-Learn: Language Model Guided RL for Solving Long Horizon Robotics Tasks