Dita: Scaling Diffusion Transformer for Generalist Vision-Language-Action Policy

• URL: http://arxiv.org/abs/2503.19757v1
• Date: Tue, 25 Mar 2025 15:19:56 GMT
• Title: Dita: Scaling Diffusion Transformer for Generalist Vision-Language-Action Policy
• Authors: Zhi Hou, Tianyi Zhang, Yuwen Xiong, Haonan Duan, Hengjun Pu, Ronglei Tong, Chengyang Zhao, Xizhou Zhu, Yu Qiao, Jifeng Dai, Yuntao Chen,
• Abstract summary: We present Dita, a scalable framework that leverages Transformer architectures to directly denoise continuous action sequences.<n>Dita employs in-context conditioning -- enabling fine-grained alignment between denoised actions and raw visual tokens from historical observations.<n>Dita effectively integrates cross-embodiment datasets across diverse camera perspectives, observation scenes, tasks, and action spaces.
• Score: 56.424032454461695
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: While recent vision-language-action models trained on diverse robot datasets exhibit promising generalization capabilities with limited in-domain data, their reliance on compact action heads to predict discretized or continuous actions constrains adaptability to heterogeneous action spaces. We present Dita, a scalable framework that leverages Transformer architectures to directly denoise continuous action sequences through a unified multimodal diffusion process. Departing from prior methods that condition denoising on fused embeddings via shallow networks, Dita employs in-context conditioning -- enabling fine-grained alignment between denoised actions and raw visual tokens from historical observations. This design explicitly models action deltas and environmental nuances. By scaling the diffusion action denoiser alongside the Transformer's scalability, Dita effectively integrates cross-embodiment datasets across diverse camera perspectives, observation scenes, tasks, and action spaces. Such synergy enhances robustness against various variances and facilitates the successful execution of long-horizon tasks. Evaluations across extensive benchmarks demonstrate state-of-the-art or comparative performance in simulation. Notably, Dita achieves robust real-world adaptation to environmental variances and complex long-horizon tasks through 10-shot finetuning, using only third-person camera inputs. The architecture establishes a versatile, lightweight and open-source baseline for generalist robot policy learning. Project Page: https://robodita.github.io.

Related papers

• DiffMoE: Dynamic Token Selection for Scalable Diffusion Transformers [86.5541501589166]
  DiffMoE is a batch-level global token pool that enables experts to access global token distributions during training. It achieves state-of-the-art performance among diffusion models on ImageNet benchmark. The effectiveness of our approach extends beyond class-conditional generation to more challenging tasks such as text-to-image generation.
  arXiv  Detail & Related papers  (2025-03-18T17:57:07Z)
• ManipGPT: Is Affordance Segmentation by Large Vision Models Enough for Articulated Object Manipulation? [17.356760351203715]
  This paper introduces ManipGPT, a framework designed to predict optimal interaction areas for articulated objects. We created a dataset of 9.9k simulated and real images to bridge the sim-to-real gap. We significantly improved part-level affordance segmentation, adapting the model's in-context segmentation capabilities to robot manipulation scenarios.
  arXiv  Detail & Related papers  (2024-12-13T11:22:01Z)
• LaVin-DiT: Large Vision Diffusion Transformer [99.98106406059333]
  LaVin-DiT is a scalable and unified foundation model designed to tackle over 20 computer vision tasks in a generative framework. We introduce key innovations to optimize generative performance for vision tasks. The model is scaled from 0.1B to 3.4B parameters, demonstrating substantial scalability and state-of-the-art performance across diverse vision tasks.
  arXiv  Detail & Related papers  (2024-11-18T12:05:27Z)
• Diffusion Transformer Policy [48.50988753948537]
  We propose a large multi-modal diffusion transformer, dubbed as Diffusion Transformer Policy, to model continuous end-effector actions.<n>By leveraging the scaling capability of transformers, the proposed approach can effectively model continuous end-effector actions across large diverse robot datasets.
  arXiv  Detail & Related papers  (2024-10-21T12:43:54Z)
• SAM-E: Leveraging Visual Foundation Model with Sequence Imitation for Embodied Manipulation [62.58480650443393]
  Segment Anything (SAM) is a vision-foundation model for generalizable scene understanding and sequence imitation. We develop a novel multi-channel heatmap that enables the prediction of the action sequence in a single pass.
  arXiv  Detail & Related papers  (2024-05-30T00:32:51Z)
• SGDViT: Saliency-Guided Dynamic Vision Transformer for UAV Tracking [12.447854608181833]
  This work presents a novel saliency-guided dynamic vision Transformer (SGDViT) for UAV tracking. The proposed method designs a new task-specific object saliency mining network to refine the cross-correlation operation. A lightweight saliency filtering Transformer further refines saliency information and increases the focus on appearance information.
  arXiv  Detail & Related papers  (2023-03-08T05:01:00Z)
• Transformers Solve the Limited Receptive Field for Monocular Depth Prediction [82.90445525977904]
  We propose TransDepth, an architecture which benefits from both convolutional neural networks and transformers. This is the first paper which applies transformers into pixel-wise prediction problems involving continuous labels.
  arXiv  Detail & Related papers  (2021-03-22T18:00:13Z)

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.