# Robotic Policy Adaptation via Weight-Space Meta-Learning

arXiv: https://arxiv.org/abs/2606.07217 · HTML: https://mlsanity.com/paper/2606.07217/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning and post-training, one of 13 directions in Vision-language-action models.
- Builds on 29 of its 35 references that are in the map.
- First posted on arXiv 5 Jun 2026.

## Placed in

- Vision-language-action models / [Reinforcement learning and post-training](https://mlsanity.com/vla/reinforcement-learning-and-post-training/)

## Builds on (most cited first, 29 of 29)

- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [PaLM-E: An Embodied Multimodal Language Model](https://mlsanity.com/paper/2303.03378/) (2023)
- [RT-1: Robotics Transformer for Real-World Control at Scale](https://mlsanity.com/paper/2212.06817/) (2022)
- [π0.5: a Vision-Language-Action Model with Open-World Generalization](https://mlsanity.com/paper/2504.16054/) (2025)
- [LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning](https://mlsanity.com/paper/2306.03310/) (2023)
- [RMA: Rapid Motor Adaptation for Legged Robots](https://mlsanity.com/paper/2107.04034/) (2021)
- [VIMA: General Robot Manipulation with Multimodal Prompts](https://mlsanity.com/paper/2210.03094/) (2022)
- [MT-Opt: Continuous Multi-Task Robotic Reinforcement Learning at Scale](https://mlsanity.com/paper/2104.08212/) (2021)
- [Task2Vec: Task Embedding for Meta-Learning](https://mlsanity.com/paper/1902.03545/) (2019)
- [SMASH: One-Shot Model Architecture Search through HyperNetworks](https://mlsanity.com/paper/1708.05344/) (2017)
- [LIBERO-PRO: Towards Robust and Fair Evaluation of Vision-Language-Action Models Beyond Memorization](https://mlsanity.com/paper/2510.03827/) (2025)
- [Using Fast Weights to Attend to the Recent Past](https://mlsanity.com/paper/1610.06258/) (2016)
- [Parameter Prediction for Unseen Deep Architectures](https://mlsanity.com/paper/2110.13100/) (2021)
- [Learning to Learn with Generative Models of Neural Network Checkpoints](https://mlsanity.com/paper/2209.12892/) (2022)
- [Prediction with Action: Visual Policy Learning via Joint Denoising Process](https://mlsanity.com/paper/2411.18179/) (2024)
- [Task-Embedded Control Networks for Few-Shot Imitation Learning](https://mlsanity.com/paper/1810.03237/) (2018)
- [Hyper-Representations as Generative Models: Sampling Unseen Neural Network Weights](https://mlsanity.com/paper/2209.14733/) (2022)
- [Hypernetworks in Meta-Reinforcement Learning](https://mlsanity.com/paper/2210.11348/) (2022)
- [Diffusion-Based Neural Network Weights Generation](https://mlsanity.com/paper/2402.18153/) (2024)
- [METAL: A Multi-Agent Framework for Chart Generation with Test-Time Scaling](https://mlsanity.com/paper/2502.17651/) (2025)
- [Drag-and-Drop LLMs: Zero-Shot Prompt-to-Weights](https://mlsanity.com/paper/2506.16406/) (2025)
- [One-Shot Imitation Learning with Invariance Matching for Robotic Manipulation](https://mlsanity.com/paper/2405.13178/) (2024)
- [Conditional LoRA Parameter Generation](https://mlsanity.com/paper/2408.01415/) (2024)
- [Recurrent Diffusion for Large-Scale Parameter Generation](https://mlsanity.com/paper/2501.11587/) (2025)
- [Make-An-Agent: A Generalizable Policy Network Generator with Behavior-Prompted Diffusion](https://mlsanity.com/paper/2407.10973/) (2024)
- [MetaVLA: Unified Meta Co-training For Efficient Embodied Adaption](https://mlsanity.com/paper/2510.05580/) (2025)
- [WARPD: World model Assisted Reactive Policy Diffusion](https://mlsanity.com/paper/2410.14040/) (2024)
- [Hyper-GoalNet: Goal-Conditioned Manipulation Policy Learning with HyperNetworks](https://mlsanity.com/paper/2512.00085/) (2025)
- [Generative NeuroEvolution for Deep Learning](https://mlsanity.com/paper/1312.5355/) (2013)

## Built on by (most cited first, 1 of 1)

- [Amortized Low-Rank Adaptation for Model-Based Reinforcement Learning](https://mlsanity.com/paper/2609.12278/) (2026)

## Abstract (arXiv)

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.