Robotic Policy Adaptation via Weight-Space Meta-Learning

• URL: http://arxiv.org/abs/2606.07217v1
• Date: Fri, 05 Jun 2026 12:29:28 GMT
• Title: Robotic Policy Adaptation via Weight-Space Meta-Learning
• Abstract summary: WIZARD sidesteps task-specific fine-tuning by generating task-specific LoRA parameters for a frozen VLA policy.<n>During meta-training, WIZARD learns to map task evidence directly to expert LoRA updates.<n>Experiments show that WIZARD improves performance by up to 2x on unseen dataset collections and up to 14x on unseen tasks.
• Score: 10.867247751424891
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision-Language-Action (VLA) models are emerging as a promising paradigm for robotic manipulation, enabling general-purpose policies trained from large corpora of demonstrations and action labels. However, adapting these models to new tasks still typically requires task-specific demonstrations, action annotations, and additional fine-tuning, making deployment costly and difficult to scale. We propose WIZARD, a weight-space meta-learning framework that sidesteps task-specific fine-tuning by generating task-specific LoRA parameters for a frozen VLA policy. Given only a language instruction and a short demonstration video, WIZARD predicts the corresponding adaptation weights in a single forward pass, without target-task action labels or test-time optimization. During meta-training, WIZARD learns to map task evidence directly to expert LoRA updates, capturing relationships between tasks in weight space. Experiments on LIBERO show that WIZARD improves performance by up to ~2x on unseen dataset collections and up to ~14x on unseen tasks. On a Franka Emika Panda, WIZARD consistently improves over a real-domain adapted baseline, showing that generated adapters provide task-level specialization beyond simulation.

Related papers

• FADA: Few-Shot Domain Adaptation via Dynamics Alignment for Humanoid Control [20.273427277208558]
  We present FADA, a three-stage Planner-Inverse Dynamics Model (Planner-IDM) framework for few-shot adaptation in humanoid control.<n>Experiments show that FADA outperforms both in-context and end-to-end adaptation baselines.
  arXiv  Detail & Related papers  (2026-06-26T16:05:10Z)
• Retrieve, Don't Retrain: Extending Vision Language Action Models to New Tasks at Test Time [48.339524522652276]
  Retrieval improves policies beyond a specific backbone, including standard VLA policies, but its effect is especially pronounced in Cosmos Policy.<n>On PushT, we study how retrieval provides a reusable high-level motion prior for cross-embodiment generalization to unseen goal angles, while on RoboTwin 2.0 our method outperforms cross-embodiment baselines on unseen tasks.
  arXiv  Detail & Related papers  (2026-06-14T06:48:01Z)
• CLARE: Continual Learning for Vision-Language-Action Models via Autonomous Adapter Routing and Expansion [9.808005698482914]
  CLARE is a framework for exemplar-free continual learning with vision-language-action models.<n>We show that CLARE achieves high performance on new tasks without catastrophic forgetting of earlier tasks.
  arXiv  Detail & Related papers  (2026-01-14T14:23:42Z)
• Learning to Reason and Navigate: Parameter Efficient Action Planning with Large Language Models [63.765846080050906]
  This paper proposes a novel parameter-efficient action planner using large language models (PEAP-LLM) to generate a single-step instruction at each location.<n>Experiments show the superiority of our proposed model on REVERIE compared to the previous state-of-the-art.
  arXiv  Detail & Related papers  (2025-05-12T12:38:20Z)
• Flex: End-to-End Text-Instructed Visual Navigation from Foundation Model Features [59.892436892964376]
  We investigate the minimal data requirements and architectural adaptations necessary to achieve robust closed-loop performance with vision-based control policies.<n>Our findings are synthesized in Flex (Fly lexically), a framework that uses pre-trained Vision Language Models (VLMs) as frozen patch-wise feature extractors.<n>We demonstrate the effectiveness of this approach on a quadrotor fly-to-target task, where agents trained via behavior cloning successfully generalize to real-world scenes.
  arXiv  Detail & Related papers  (2024-10-16T19:59:31Z)
• From Instance Training to Instruction Learning: Task Adapters Generation from Instructions [29.452006810725184]
  This paper focuses on simulating human learning to address the shortcomings of instance training.<n>We introduce Task Adapters Generation from Instructions (TAGI), which automatically constructs the task-specific model.<n>We evaluate TAGI on the Super-Natural Instructions and P3 datasets.
  arXiv  Detail & Related papers  (2024-06-18T08:14:28Z)
• Task-Distributionally Robust Data-Free Meta-Learning [99.56612787882334]
  Data-Free Meta-Learning (DFML) aims to efficiently learn new tasks by leveraging multiple pre-trained models without requiring their original training data. For the first time, we reveal two major challenges hindering their practical deployments: Task-Distribution Shift ( TDS) and Task-Distribution Corruption (TDC)
  arXiv  Detail & Related papers  (2023-11-23T15:46:54Z)
• Plan, Eliminate, and Track -- Language Models are Good Teachers for Embodied Agents [99.17668730578586]
  Pre-trained large language models (LLMs) capture procedural knowledge about the world. Plan, Eliminate, and Track (PET) framework translates a task description into a list of high-level sub-tasks. PET framework leads to a significant 15% improvement over SOTA for generalization to human goal specifications.
  arXiv  Detail & Related papers  (2023-05-03T20:11:22Z)
• Task Adaptive Parameter Sharing for Multi-Task Learning [114.80350786535952]
  Adaptive Task Adapting Sharing (TAPS) is a method for tuning a base model to a new task by adaptively modifying a small, task-specific subset of layers. Compared to other methods, TAPS retains high accuracy on downstream tasks while introducing few task-specific parameters. We evaluate our method on a suite of fine-tuning tasks and architectures (ResNet, DenseNet, ViT) and show that it achieves state-of-the-art performance while being simple to implement.
  arXiv  Detail & Related papers  (2022-03-30T23:16:07Z)

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.