pith. sign in

arxiv: 2506.16211 · v1 · pith:2QIT3AKDnew · submitted 2025-06-19 · 💻 cs.RO

ControlVLA: Few-shot Object-centric Adaptation for Pre-trained Vision-Language-Action Models

classification 💻 cs.RO
keywords controlvlademonstrationsmanipulationobject-centricpre-trainedtasksachieveexperiments
0
0 comments X
read the original abstract

Learning real-world robotic manipulation is challenging, particularly when limited demonstrations are available. Existing methods for few-shot manipulation often rely on simulation-augmented data or pre-built modules like grasping and pose estimation, which struggle with sim-to-real gaps and lack extensibility. While large-scale imitation pre-training shows promise, adapting these general-purpose policies to specific tasks in data-scarce settings remains unexplored. To achieve this, we propose ControlVLA, a novel framework that bridges pre-trained VLA models with object-centric representations via a ControlNet-style architecture for efficient fine-tuning. Specifically, to introduce object-centric conditions without overwriting prior knowledge, ControlVLA zero-initializes a set of projection layers, allowing them to gradually adapt the pre-trained manipulation policies. In real-world experiments across 6 diverse tasks, including pouring cubes and folding clothes, our method achieves a 76.7% success rate while requiring only 10-20 demonstrations -- a significant improvement over traditional approaches that require more than 100 demonstrations to achieve comparable success. Additional experiments highlight ControlVLA's extensibility to long-horizon tasks and robustness to unseen objects and backgrounds.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 9 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. X-Tokenizer: A Multimodal Action Tokenizer for Vision-Language-Action Pretraining

    cs.CV 2026-06 unverdicted novelty 7.0

    X-Tokenizer creates semantic action tokens via asymmetric residual quantization and contrastive pretraining on large trajectory data, outperforming prior methods like FAST on robotic tasks.

  2. MaskWAM: Unifying Mask Prompting and Prediction for World-Action Models

    cs.CV 2026-06 unverdicted novelty 6.0

    MaskWAM unifies mask prompting and prediction in world-action models via Mixture of Transformers to improve robotic policy generalization on language-ambiguous tasks.

  3. ELAN4D: Embodiment-Centric 4D Supervision for Vision-Language-Action Models via Plug-and-Play Adaptation

    cs.RO 2026-05 unverdicted novelty 6.0

    ELAN4D introduces plug-and-play 4D keypoint track supervision from forward kinematics to enhance VLA policy generalization in robotic manipulation tasks.

  4. TMRL: Diffusion Timestep-Modulated Pretraining Enables Exploration for Efficient Policy Finetuning

    cs.RO 2026-05 unverdicted novelty 6.0

    TMRL bridges behavioral cloning pretraining and RL finetuning via diffusion noise and timestep modulation to enable controlled exploration, improving sample efficiency and enabling real-world robot training in under one hour.

  5. Breaking Lock-In: Preserving Steerability under Low-Data VLA Post-Training

    cs.RO 2026-04 unverdicted novelty 6.0

    DeLock mitigates lock-in in low-data VLA post-training via visual grounding preservation and test-time contrastive prompt guidance, outperforming baselines across eight evaluations while matching data-heavy generalist...

  6. OFlow: Injecting Object-Aware Temporal Flow Matching for Robust Robotic Manipulation

    cs.RO 2026-04 unverdicted novelty 6.0

    OFlow unifies temporal foresight and object-aware reasoning inside a shared latent space via flow matching to improve VLA robustness in robotic manipulation under distribution shifts.

  7. FOCA: Future-Oriented Conditioning for Data-Efficient Vision-Language-Action Adaptation

    cs.CV 2026-06 unverdicted novelty 5.0

    FOCA improves few-shot VLA adaptation by explicitly predicting future interaction embeddings and implicitly aligning to goal observations, yielding up to 26% gains on real robots with only 20 demonstrations.

  8. CoEnv: Driving Embodied Multi-Agent Collaboration via Compositional Environment

    cs.RO 2026-04 unverdicted novelty 5.0

    CoEnv introduces a compositional environment that integrates real and simulated spaces for multi-agent robotic collaboration, using real-to-sim reconstruction, VLM action synthesis, and validated sim-to-real transfer ...

  9. SlotVLA: Towards Modeling of Object-Relation Representations in Robotic Manipulation

    cs.RO 2025-11 unverdicted novelty 5.0

    SlotVLA uses slot attention to model object-relation representations for multitask robotic manipulation, reducing visual tokens while achieving competitive generalization on the new LIBERO+ benchmark.