Fast-dVLA: Accelerating Discrete Diffusion VLA to Real-Time Performance

• URL: http://arxiv.org/abs/2603.25661v2
• Date: Fri, 27 Mar 2026 11:46:16 GMT
• Title: Fast-dVLA: Accelerating Discrete Diffusion VLA to Real-Time Performance
• Abstract summary: Some advanced finetuning methods with auxiliary training objectives can improve performance and reduce the number of convergence steps.<n>We propose a novel approach to address the challenge that pretrained VLA models often fail to effectively improve performance and reduce adaptation costs during standard supervised finetuning.
• Score: 47.605498477489306
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: This paper proposes a novel approach to address the challenge that pretrained VLA models often fail to effectively improve performance and reduce adaptation costs during standard supervised finetuning (SFT). Some advanced finetuning methods with auxiliary training objectives can improve performance and reduce the number of convergence steps. However, they typically incur significant computational overhead due to the additional losses from auxiliary tasks. To simultaneously achieve the enhanced capabilities of auxiliary training with the simplicity of standard SFT, we decouple the two objectives of auxiliary task training within the parameter space, namely, enhancing general capabilities and fitting task-specific action distributions. To deliver this goal, we only need to train the model to converge on a small-scale task set using two distinct training strategies. The difference between the resulting model parameters can then be interpreted as capability vectors provided by auxiliary tasks. These vectors are then merged with pretrained parameters to form a capability-enhanced meta model. Moreover, when standard SFT is augmented with a lightweight orthogonal regularization loss, the merged model attains performance comparable to auxiliary finetuned baselines with reduced computational overhead. Experimental results demonstrate that this approach is highly effective across diverse robot tasks. Project page: https://chris1220313648.github.io/Fast-dVLA/

Related papers

• CapVector: Learning Transferable Capability Vectors in Parametric Space for Vision-Language-Action Models [42.639416481955344]
  Some advanced finetuning methods with auxiliary training objectives can improve performance and reduce the number of convergence steps.<n>We propose a novel approach to address the challenge that pretrained VLA models often fail to effectively improve performance and reduce adaptation costs during standard supervised finetuning.
  arXiv  Detail & Related papers  (2026-05-11T17:41:54Z)
• VADF: Vision-Adaptive Diffusion Policy Framework for Efficient Robotic Manipulation [26.316939305877145]
  Diffusion policies are becoming mainstream in robotic manipulation but suffer from hard negative class imbalance due to uniform sampling and lack of sample difficulty awareness.<n>We propose VADF (Vision-Adaptive Diffusion Policy Framework), a vision-driven dual-adaptive framework that significantly achieves early success in inference.
  arXiv  Detail & Related papers  (2026-04-17T10:56:59Z)
• FAAR: Efficient Frequency-Aware Multi-Task Fine-Tuning via Automatic Rank Selection [13.382659738606511]
  This paper introduces Frequency-Aware and Automatic Rank (FAAR) for efficient MTL fine-tuning.<n>FAAR reduces the number of parameters by up to 9 times compared to traditional MTL fine-tuning.
  arXiv  Detail & Related papers  (2026-03-20T18:24:06Z)
• Basis-Oriented Low-rank Transfer for Few-Shot and Test-Time Adaptation [10.804106052326402]
  Adapting large pre-trained models to unseen tasks under tight data and compute budgets remains challenging.<n>We propose BOLT, a framework that reuses existing fine-tuned models and adapts within that subspace.<n>Our results show that constraining adaptation to a task-informed subspace provides an effective alternative for unseen-task transfer.
  arXiv  Detail & Related papers  (2025-12-02T06:00:16Z)
• Align-Then-stEer: Adapting the Vision-Language Action Models through Unified Latent Guidance [63.33213516925946]
  We introduce textbfAlign-Then-stEer (textttATE), a novel, data-efficient, and plug-and-play adaptation framework.<n>Our work presents a general and lightweight solution that greatly enhances the practicality of deploying VLA models to new robotic platforms and tasks.
  arXiv  Detail & Related papers  (2025-09-02T07:51:59Z)
• PYRA: Parallel Yielding Re-Activation for Training-Inference Efficient Task Adaptation [61.57833648734164]
  We propose a novel Parallel Yielding Re-Activation (PYRA) method for training-inference efficient task adaptation. PYRA outperforms all competing methods under both low compression rate and high compression rate.
  arXiv  Detail & Related papers  (2024-03-14T09:06:49Z)
• Parameter and Computation Efficient Transfer Learning for Vision-Language Pre-trained Models [79.34513906324727]
  In this paper, we aim at parameter and efficient transfer learning (PCETL) for vision-language pre-trained models. We propose a novel dynamic architecture skipping (DAS) approach towards effective PCETL.
  arXiv  Detail & Related papers  (2023-09-04T09:34:33Z)
• DL-DRL: A double-level deep reinforcement learning approach for large-scale task scheduling of multi-UAV [65.07776277630228]
  We propose a double-level deep reinforcement learning (DL-DRL) approach based on a divide and conquer framework (DCF) Particularly, we design an encoder-decoder structured policy network in our upper-level DRL model to allocate the tasks to different UAVs. We also exploit another attention based policy network in our lower-level DRL model to construct the route for each UAV, with the objective to maximize the number of executed tasks.
  arXiv  Detail & Related papers  (2022-08-04T04:35:53Z)

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.