DIAL: Decoupling Intent and Action via Latent World Modeling for End-to-End VLA

• URL: http://arxiv.org/abs/2603.29844v1
• Date: Tue, 31 Mar 2026 15:02:27 GMT
• Title: DIAL: Decoupling Intent and Action via Latent World Modeling for End-to-End VLA
• Abstract summary: 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.
• Score: 72.9197085473598
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The development of Vision-Language-Action (VLA) models has been significantly accelerated by pre-trained Vision-Language Models (VLMs). However, most existing end-to-end VLAs treat the VLM primarily as a multimodal encoder, directly mapping vision-language features to low-level actions. This paradigm underutilizes the VLM's potential in high-level decision making and introduces training instability, frequently degrading its rich semantic representations. To address these limitations, we introduce DIAL, a framework bridging high-level decision making and low-level motor execution through a differentiable latent intent bottleneck. Specifically, a VLM-based System-2 performs latent world modeling by synthesizing latent visual foresight within the VLM's native feature space; this foresight explicitly encodes intent and serves as the structural bottleneck. A lightweight System-1 policy then decodes this predicted intent together with the current observation into precise robot actions via latent inverse dynamics. To ensure optimization stability, we employ a two-stage training paradigm: a decoupled warmup phase where System-2 learns to predict latent futures while System-1 learns motor control under ground-truth future guidance within a unified feature space, followed by seamless end-to-end joint optimization. This enables action-aware gradients to refine the VLM backbone in a controlled manner, preserving pre-trained knowledge. Extensive experiments on the RoboCasa GR1 Tabletop benchmark show that DIAL establishes a new state-of-the-art, achieving superior performance with 10x fewer demonstrations than prior methods. Furthermore, by leveraging heterogeneous human demonstrations, DIAL learns physically grounded manipulation priors and exhibits robust zero-shot generalization to unseen objects and novel configurations during real-world deployment on a humanoid robot.

Related papers

• InternVLA-A1.5: Unifying Understanding, Latent Foresight, and Action for Compositional Generalization [81.9611022323101]
  We present InternVLA-A1.5, which builds the policy on a native VLM backbone that keeps training on VQA and subtask prediction, and attaches a lightweight unified expert for continuous action generation.<n>Future prediction is recast as a latent-querying problem, where a small set of learnable foresight tokens condenses the task-relevant future into a compact latent code under the supervision of a frozen pretrained video generation model.<n>Pretrained on 1.2M robot episodes and 3M multimodal samples, InternVLA-A1.5 achieves the best overall results on all six simulation benchmarks.
  arXiv  Detail & Related papers  (2026-07-06T12:25:30Z)
• Two Bridges, One Pathway: From VLMs to Generalizable VLAs with Embodied Trajectory-Coupled Data [93.7685703383343]
  Vision-language models (VLMs) are powerful general-purpose reasoners, yet converting them into robot control policies is surprisingly difficult.<n>We argue that this gap can be bridged gradually with the right intermediate data.<n>We show that mixing task-relevant out-of-distribution ETC data with a small amount of action data enables the model to generalize to novel visual-language conditions.
  arXiv  Detail & Related papers  (2026-06-07T08:57:51Z)
• $M^2$-VLA: Boosting Vision-Language Models for Generalizable Manipulation via Layer Mixture and Meta-Skills [25.17452377052361]
  Current Vision-Language-Action (VLA) models predominantly rely on end-to-end fine-tuning.<n>We propose $M2$-VLA, which demonstrates that a generalized VLM is able to serve as a powerful backbone for robotic manipulation directly.<n>To overcome this, we introduce the Mixture of Layers (MoL) strategy that selectively extracts task-critical information from dense semantic features.
  arXiv  Detail & Related papers  (2026-04-27T08:44:12Z)
• MMaDA-VLA: Large Diffusion Vision-Language-Action Model with Unified Multi-Modal Instruction and Generation [47.2088748094605]
  Vision-Language-Action (VLA) models aim to control robots for manipulation from visual observations and natural-language instructions.<n>We present MMaDA-VLA, a fully native pre-trained large diffusion VLA model that unifies multi-modal understanding and generation in a single framework.
  arXiv  Detail & Related papers  (2026-03-26T12:55:51Z)
• Self-Correcting VLA: Online Action Refinement via Sparse World Imagination [55.982504915794514]
  We propose Self-Correcting VLA (SC-VLA), which achieve self-improvement by intrinsically guiding action refinement through sparse imagination.<n>SC-VLA achieve state-of-the-art performance, yielding the highest task throughput with 16% fewer steps and a 9% higher success rate than the best-performing baselines.
  arXiv  Detail & Related papers  (2026-02-25T06:58:06Z)
• ST4VLA: Spatially Guided Training for Vision-Language-Action Models [80.35847468618276]
  Large vision-language models (VLMs) excel at multimodal understanding but fall short when extended to embodied tasks.<n>We introduce ST4VLA, a dual-system Vision-Language-Action framework to align action learning with spatial priors inVLMs.<n> Empirically, ST4VLA achieves substantial improvements over vanilla VLA, with performance increasing from 66.1 -> 84.6 on Google Robot and from 54.7 -> 73.2 on WidowX Robot.
  arXiv  Detail & Related papers  (2026-02-10T18:59:17Z)
• On-the-Fly VLA Adaptation via Test-Time Reinforcement Learning [61.38273866396522]
  Vision-Language-Action models have emerged as a powerful paradigm for general-purpose robot learning.<n>Current methods remain unsuitable for challenging simulated- or physical-world deployments.<n>We introduce a Test-Time Reinforcement Learning for VLAs framework that enables on-the-fly policy adaptation during inference.
  arXiv  Detail & Related papers  (2026-01-11T01:51:30Z)
• mimic-video: Video-Action Models for Generalizable Robot Control Beyond VLAs [5.109732854501585]
  We introduce mimic-video, a novel Video-Action Model (VAM) that pairs a pretrained Internet-scale video model with a flow matching-based action decoder conditioned on its latent representations.<n>Our approach achieves state-of-the-art performance on simulated and real-world robotic manipulation tasks, improving sample efficiency by 10x and convergence speed by 2x compared to traditional VLA architectures.
  arXiv  Detail & Related papers  (2025-12-17T18:47:31Z)
• Bridge Thinking and Acting: Unleashing Physical Potential of VLM with Generalizable Action Expert [60.88976842557026]
  Vision-Language Models (VLM) have demonstrated impressive planning and reasoning capabilities.<n>Recent dual-system approaches attempt to decouple "thinking" from "acting"<n>We introduce a framework centered around a generalizable action expert.
  arXiv  Detail & Related papers  (2025-10-04T18:33:27Z)
• dVLA: Diffusion Vision-Language-Action Model with Multimodal Chain-of-Thought [66.78110237549087]
  Vision-Language-Action (VLA) models are emerging as a next-generation paradigm for robotics.<n>We introduce dVLA, a diffusion-based VLA that unifies visual perception, language reasoning, and robotic control in a single system.
  arXiv  Detail & Related papers  (2025-09-30T02:36:11Z)

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.