HAF: Adapting Generalist VLAs to Humanoid Whole-Body Loco-manipulation via Hierarchical Action Flow and Spectral Latent RL

• URL: http://arxiv.org/abs/2608.16837v1
• Date: Mon, 17 Aug 2026 17:22:33 GMT
• Title: HAF: Adapting Generalist VLAs to Humanoid Whole-Body Loco-manipulation via Hierarchical Action Flow and Spectral Latent RL
• Abstract summary: We introduce HAF (Humanoid Adaptation Framework) to transfer off-the-shelf generalist VLA foundation models to humanoid whole-body loco-manipulation.<n> HAF-VLA is a hierarchical action-flow generator built on a pretrained flow-matching VLA.<n>HaF-Steer is a latent offline-to-online RL pipeline that leverages flow-matching invertibility and DCT-based dimensionality reduction.
• Score: 56.014913786737104
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Humanoid robots hold great promise as general-purpose agents in human-centered environments, yet generalist vision-language-action (VLA) foundation models are not readily applicable to humanoid whole-body loco-manipulation. The high dimensionality and interdependence of humanoid motions make it challenging for conventional single-stage VLA architectures to coordinate locomotion, waist posture, and dual-arm manipulation effectively. Moreover, policies trained through offline behavior cloning can remain suboptimal during real-world deployment. Although online reinforcement learning can refine policies through real-world interaction, directly tuning large VLA backbones demands excessive computation and may introduce safety risks during real-robot exploration. To address these bottlenecks, we introduce HAF (Humanoid Adaptation Framework), a two-part framework consisting of HAF-VLA and HAF-Steer that transfers off-the-shelf generalist VLA foundation models to humanoid whole-body loco-manipulation. HAF-VLA is a hierarchical action-flow generator built on a pretrained flow-matching VLA. It splits full-body action denoising into three sequential stages with stage embeddings and cross-stage KV caches that retain kinematic dependencies, avoiding incoherent whole-body actions from one-shot generation. On top of the frozen HAF-VLA, HAF-Steer is a latent offline-to-online RL pipeline that leverages flow-matching invertibility and DCT-based dimensionality reduction to restrict RL optimization to a compact noise subspace and train a regularized SAC policy. This avoids updating the large VLA backbone and enables efficient real-world policy refinement. Evaluated on seven real-world humanoid loco-manipulation tasks, HAF surpasses vanilla single-stage VLA baselines and improves whole-body coordination and task performance. Project website: https://grange007.github.io/HAF .

Related papers

• StructRL: Structured Action-Space Exploration for Flow-Based VLAs [81.43458629514139]
  Flow-based Vision-Language-Action models are now widely used for continuous robotic manipulation.<n>Online reinforcement learning is emerging as a key technique for adapting them to new tasks.<n>We show that effective robot exploration calls for structured noise: temporally smooth and scaled differently across action groups.
  arXiv  Detail & Related papers  (2026-08-15T09:31:29Z)
• FlowDAgger: Human-in-the-Loop Adaptation of Generative Robot Policies in Latent Space [15.351960740594201]
  FlowDAgger is a sample- and compute-efficient method for adapting frozen generative robot policies from human interventions in latent space.<n>Our key idea is action inversion: each human expert action is mapped to the noise that would have produced it under the frozen base policy.<n>We evaluate FlowDAgger in simulation and on real-world bimanual and single-arm manipulation.
  arXiv  Detail & Related papers  (2026-07-09T19:07:33Z)
• Object-Centric Residual RL for Zero-Shot Sim-to-Real VLA Enhancement [52.93267443851685]
  Vision-Language-Action (VLA) models can generalize across diverse manipulation tasks.<n>Can a reinforcement learning policy trained purely in simulation improve the robustness of real-world VLAs zero-shot?<n>We propose an object-centric residual RL framework that refines VLA actions using object poses.
  arXiv  Detail & Related papers  (2026-06-17T11:36:54Z)
• HOIST: Humanoid Optimization with Imitation and Sample-efficient Tuning for Manipulating Suspended Loads [26.63242922377076]
  We present HOIST-Humanoid optimized with Imitation and Sample-efficient Tuning for manipulating suspended loads.<n> HOIST first finetunes a high-level vision--action (VLA) policy from virtual-reality (VR) teleoperation demonstrations.<n>It then uses VLA rollouts and iterative batched RL to improve placement accuracy and stopping behavior.
  arXiv  Detail & Related papers  (2026-05-29T18:33:32Z)
• Jump-Start Reinforcement Learning with Vision-Language-Action Regularization [1.2599533416395767]
  Reinforcement learning (RL) enables high-frequency, closed-loop control for robotic manipulation.<n>Current limitations hinder their direct use in fast and precise manipulation.<n>We propose Vision-Language-Action Jump-Starting (VLAJS) to improve exploration and learning efficiency.
  arXiv  Detail & Related papers  (2026-04-15T11:17:54Z)
• DIAL: Decoupling Intent and Action via Latent World Modeling for End-to-End VLA [72.9197085473598]
  DIAL is a framework bridging high-level decision making and low-level motor execution through a differentiable latent intent bottleneck.<n>A VLM-based System-2 performs latent world modeling by latent visual foresight within the VLM's native feature space.<n>A lightweight System-1 policy then decodes this predicted intent together with the current observation into precise robot actions.
  arXiv  Detail & Related papers  (2026-03-31T15:02:27Z)
• ULTRA: Unified Multimodal Control for Autonomous Humanoid Whole-Body Loco-Manipulation [55.467742403416175]
  We introduce a physics-driven neural algorithm that translates large-scale motion capture to humanoid embodiments.<n>We learn a unified multimodal controller that supports both dense references and sparse task specifications.<n>Results show that ULTRA generalizes to autonomous, goal-conditioned whole-body loco-manipulation from egocentric perception.
  arXiv  Detail & Related papers  (2026-03-03T18:59:29Z)
• LiLo-VLA: Compositional Long-Horizon Manipulation via Linked Object-Centric Policies [54.150202739999806]
  LiLo-VLA is a modular framework capable of zero-shot modularity to novel long-horizon tasks without ever being trained on them.<n>We introduce a 21-task simulation benchmark consisting of two challenging suites: LIBERO-Long++ and Ultra-Long.<n>In these simulations, LiLo-VLA achieves a 69% average success rate, outperforming Pi0.5 by 41% and OpenVLA-OFT by 67%.
  arXiv  Detail & Related papers  (2026-02-25T03:33:39Z)
• Human-assisted Robotic Policy Refinement via Action Preference Optimization [26.144183856600687]
  Action Preference Optimization (APO) is a method designed to refine Vision-Language-Action (VLA) models by human-assisted preference alignment.<n>To solve this, APO proposes an adaptive reweighting algorithm with binary desirability signals derived from interaction.<n>Experiments conducted in simulation and real-world scenarios prove superior generalization and robustness.
  arXiv  Detail & Related papers  (2025-06-08T13:14:18Z)

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.