VLGOR: Visual-Language Knowledge Guided Offline Reinforcement Learning for Generalizable Agents

• URL: http://arxiv.org/abs/2603.22892v1
• Date: Tue, 24 Mar 2026 07:40:30 GMT
• Title: VLGOR: Visual-Language Knowledge Guided Offline Reinforcement Learning for Generalizable Agents
• Abstract summary: Large Language Models (LLMs) and Reinforcement Learning (RL) enable agents to interpret language instructions more effectively for task execution.<n>We propose Visual-Language Knowledge-Guided Offline Reinforcement Learning (VLGOR), a framework that integrates visual and language knowledge to generate imaginary rollouts.<n> Experiments on robotic manipulation benchmarks demonstrate that VLGOR significantly improves performance on unseen tasks requiring novel optimal policies, achieving a success rate over 24% higher than the baseline methods.
• Score: 14.848584432075285
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Combining Large Language Models (LLMs) with Reinforcement Learning (RL) enables agents to interpret language instructions more effectively for task execution. However, LLMs typically lack direct perception of the physical environment, which limits their understanding of environmental dynamics and their ability to generalize to unseen tasks. To address this limitation, we propose Visual-Language Knowledge-Guided Offline Reinforcement Learning (VLGOR), a framework that integrates visual and language knowledge to generate imaginary rollouts, thereby enriching the interaction data. The core premise of VLGOR is to fine-tune a vision-language model to predict future states and actions conditioned on an initial visual observation and high-level instructions, ensuring that the generated rollouts remain temporally coherent and spatially plausible. Furthermore, we employ counterfactual prompts to produce more diverse rollouts for offline RL training, enabling the agent to acquire knowledge that facilitates following language instructions while grounding in environments based on visual cues. Experiments on robotic manipulation benchmarks demonstrate that VLGOR significantly improves performance on unseen tasks requiring novel optimal policies, achieving a success rate over 24% higher than the baseline methods.

Related papers

• Seeing Beyond Words: Self-Supervised Visual Learning for Multimodal Large Language Models [53.06230963851451]
  JARVIS is a JEPA-inspired framework for self-supervised visual enhancement in MLLMs.<n>We introduce JARVIS, a JEPA-inspired framework for self-supervised visual enhancement in MLLMs.
  arXiv  Detail & Related papers  (2025-12-17T19:01:34Z)
• Semore: VLM-guided Enhanced Semantic Motion Representations for Visual Reinforcement Learning [11.901989132359676]
  We introduce Enhanced Semantic Motion Representations (Semore), a new VLM-based framework for visual reinforcement learning (RL)<n>Semore simultaneously extract semantic and motion representations through a dual-path backbone from the RGB flows.<n>Our method exhibits efficient and adaptive ability compared to state-of-art methods.
  arXiv  Detail & Related papers  (2025-12-04T16:54:41Z)
• VIPER: Visual Perception and Explainable Reasoning for Sequential Decision-Making [17.313485392764353]
  VIPER is a novel framework for multimodal instruction-based planning.<n>It integrates VLM-based perception with LLM-based reasoning.<n>We show that VIPER significantly outperforms state-of-the-art visual instruction-based planners.
  arXiv  Detail & Related papers  (2025-03-19T11:05:42Z)
• The Synergy of LLMs & RL Unlocks Offline Learning of Generalizable Language-Conditioned Policies with Low-fidelity Data [50.544186914115045]
  TEDUO is a novel training pipeline for offline language-conditioned policy learning in symbolic environments.<n>Our approach harnesses large language models (LLMs) in a dual capacity: first, as automatization tools augmenting offline datasets with richer annotations, and second, as generalizable instruction-following agents.
  arXiv  Detail & Related papers  (2024-12-09T18:43:56Z)
• Visual Grounding for Object-Level Generalization in Reinforcement Learning [35.39214541324909]
  Generalization is a pivotal challenge for agents following natural language instructions. We leverage a vision-language model (VLM) for visual grounding and transfer its vision-language knowledge into reinforcement learning. We show that our intrinsic reward significantly improves performance on challenging skill learning.
  arXiv  Detail & Related papers  (2024-08-04T06:34:24Z)
• Vision-Language Models Provide Promptable Representations for Reinforcement Learning [67.40524195671479]
  We propose a novel approach that uses the vast amounts of general and indexable world knowledge encoded in vision-language models (VLMs) pre-trained on Internet-scale data for embodied reinforcement learning (RL) We show that our approach can use chain-of-thought prompting to produce representations of common-sense semantic reasoning, improving policy performance in novel scenes by 1.5 times.
  arXiv  Detail & Related papers  (2024-02-05T00:48:56Z)
• VILA: On Pre-training for Visual Language Models [74.08039416548209]
  We study the design options for VLM pre-training through step-by-step controllable comparisons. We build VILA, a Visual Language model family that consistently outperforms the state-of-the-art models.
  arXiv  Detail & Related papers  (2023-12-12T18:58:18Z)
• LMRL Gym: Benchmarks for Multi-Turn Reinforcement Learning with Language Models [56.25156596019168]
  This paper introduces the LMRL-Gym benchmark for evaluating multi-turn RL for large language models (LLMs) Our benchmark consists of 8 different language tasks, which require multiple rounds of language interaction and cover a range of tasks in open-ended dialogue and text games.
  arXiv  Detail & Related papers  (2023-11-30T03:59:31Z)
• From Language Modeling to Instruction Following: Understanding the Behavior Shift in LLMs after Instruction Tuning [63.63840740526497]
  We investigate how instruction tuning adjusts pre-trained models with a focus on intrinsic changes. The impact of instruction tuning is then studied by comparing the explanations derived from the pre-trained and instruction-tuned models. Our findings reveal three significant impacts of instruction tuning.
  arXiv  Detail & Related papers  (2023-09-30T21:16:05Z)
• Context-Aware Prompt Tuning for Vision-Language Model with Dual-Alignment [15.180715595425864]
  We introduce a novel method to improve the prompt learning of vision-language models by incorporating pre-trained large language models (LLMs) With DuAl-PT, we propose to learn more context-aware prompts, benefiting from both explicit and implicit context modeling. Empirically, DuAl-PT achieves superior performance on 11 downstream datasets on few-shot recognition and base-to-new generalization.
  arXiv  Detail & Related papers  (2023-09-08T06:51:15Z)
• Improving Policy Learning via Language Dynamics Distillation [87.27583619910338]
  We propose Language Dynamics Distillation (LDD), which pretrains a model to predict environment dynamics given demonstrations with language descriptions. We show that language descriptions in demonstrations improve sample-efficiency and generalization across environments.
  arXiv  Detail & Related papers  (2022-09-30T19:56:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.