Pith. sign in

REVIEW 17 cited by

Diffusion Transformer Policy

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2410.15959 v6 pith:FR7IAAAW submitted 2024-10-21 cs.RO cs.CV

Diffusion Transformer Policy

classification cs.RO cs.CV
keywords actiontransformerdiffusionlargecalvincontinuousdiversemodel
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Recent large vision-language-action models pretrained on diverse robot datasets have demonstrated the potential for generalizing to new environments with a few in-domain data. However, those approaches usually predict individual discretized or continuous action by a small action head, which limits the ability in handling diverse action spaces. In contrast, we model the continuous action sequence with a large multi-modal diffusion transformer, dubbed as Diffusion Transformer Policy, in which we directly denoise action chunks by a large transformer model rather than a small action head for action embedding. By leveraging the scaling capability of transformers, the proposed approach can effectively model continuous end-effector actions across large diverse robot datasets, and achieve better generalization performance. Extensive experiments demonstrate the effectiveness and generalization of Diffusion Transformer Policy on Maniskill2, Libero, Calvin and SimplerEnv, as well as the real-world Franka arm, achieving consistent better performance on Real-to-Sim benchmark SimplerEnv, real-world Franka Arm and Libero compared to OpenVLA and Octo. Specifically, without bells and whistles, the proposed approach achieves state-of-the-art performance with only a single third-view camera stream in the Calvin task ABC->D, improving the average number of tasks completed in a row of 5 to 3.6, and the pretraining stage significantly facilitates the success sequence length on the Calvin by over 1.2.

discussion (0)

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

Forward citations

Cited by 17 Pith papers

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

  1. Test-time Adversarial Takeover: A Real-time Hijacking Interface against Robotic Diffusion Policies

    cs.RO 2026-06 unverdicted novelty 8.0

    TAKO demonstrates real-time adversarial takeover of robotic diffusion policies via reusable universal patches on visual inputs, achieving 100% success in steering attacker-chosen trajectories across multiple tasks, en...

  2. RoboFlow4D: A Lightweight Flow World Model Toward Real-Time Flow-Guided Robotic Manipulation

    cs.RO 2026-05 unverdicted novelty 7.0

    RoboFlow4D is an end-to-end lightweight flow world model that predicts multi-frame 3D flows from visual observations and textual instructions to provide explicit planning for real-time robotic manipulation.

  3. CF-VLA: Efficient Coarse-to-Fine Action Generation for Vision-Language-Action Policies

    cs.CV 2026-04 unverdicted novelty 7.0

    CF-VLA uses a coarse initialization over endpoint velocity followed by single-step refinement to achieve strong performance with low inference steps on CALVIN, LIBERO, and real-robot tasks.

  4. PriGo: Test-Time Primitive Guidance to Diffusion and Flow Policies for Adaptive Robotic Manipulation

    cs.RO 2026-07 conditional novelty 6.0

    A lightweight primitive classifier and differentiable guidance mechanism improve pretrained diffusion and flow manipulation policies by 3–7 points at test time without retraining.

  5. Hierarchical Policy Learning via Spectral Decomposition

    cs.RO 2026-06 unverdicted novelty 6.0

    Causal Spectral Policy decomposes actions spectrally into coarse motion from obs/language and conditional fine corrections, outperforming baselines on precision manipulation tasks.

  6. SSI-Policy: Learning Structured Scene Interfaces for Vision-Language Robotic Manipulation

    cs.RO 2026-06 unverdicted novelty 6.0

    SSI-Policy learns a robot-agnostic RGB-only scene interface from video to improve vision-language manipulation policies by 15% on LIBERO with only 10 demos per task.

  7. SSI-Policy: Learning Structured Scene Interfaces for Vision-Language Robotic Manipulation

    cs.RO 2026-06 unverdicted novelty 6.0

    SSI-Policy uses an RGB-only Structured Scene Interface to improve LIBERO benchmark performance by nearly 15% with only 10 demonstrations per task compared to prior methods.

  8. OpenEAI-Platform: An Open-source Embodied Artificial Intelligence Hardware-Software Unified Platform

    cs.RO 2026-06 conditional novelty 6.0

    OpenEAI-Platform delivers an open-source low-cost robotic arm and VLA model that outperforms commercial arms and matches large pretrained baselines on four real-world manipulation tasks using limited open data.

  9. Continuous Reasoning for Vision-Language-Action

    cs.RO 2026-05 unverdicted novelty 6.0

    Continuous Reasoning for VLA introduces a shared Gaussian latent for continuous thoughts, trained with self-verification to improve action prediction on LIBERO-PRO and real robots.

  10. Block-wise Adaptive Caching for Accelerating Diffusion Policy

    cs.AI 2025-06 unverdicted novelty 6.0

    BAC accelerates transformer-based Diffusion Policy up to 3x by block-level adaptive feature caching using an Adaptive Caching Scheduler and Bubbling Union Algorithm to control error propagation.

  11. Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success

    cs.RO 2025-02 accept novelty 6.0

    OpenVLA-OFT fine-tuning boosts LIBERO success rate from 76.5% to 97.1%, speeds action generation 26x, and outperforms baselines on real bimanual dexterous tasks.

  12. CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulation

    cs.RO 2024-11 unverdicted novelty 6.0

    CogACT is a new VLA model that uses a conditioned diffusion action transformer to achieve over 35% higher average success rates than OpenVLA in simulation and 55% in real-robot experiments while generalizing to new ro...

  13. FlowMaps: Modeling Long-Term Multimodal Object Dynamics with Flow Matching

    cs.RO 2026-06 unverdicted novelty 5.0

    FlowMaps is a latent flow matching model that estimates multimodal distributions over future 3D object locations conditioned on past interactions to improve dynamic object navigation.

  14. IMPACT: Learning Internal-Model Predictive Control for Forceful Robotic Manipulation

    cs.RO 2026-06 unverdicted novelty 5.0

    IMPACT decouples forceful manipulation into task-planning and internal-model predictive control, claiming higher success rates, better generalization to unseen weights, and improved safety and energy efficiency in sim...

  15. World-Value-Action Model: Implicit Planning for Vision-Language-Action Systems

    cs.RO 2026-04 unverdicted novelty 5.0

    The World-Value-Action model enables implicit planning for VLA systems by performing inference over a learned latent representation of high-value future trajectories instead of direct action prediction.

  16. RESample: A Robust Data Augmentation Framework via Exploratory Sampling for Robotic Manipulation

    cs.RO 2025-10 unverdicted novelty 5.0

    RESample uses exploratory sampling guided by a lightweight Coverage Function to expand VLA training data coverage, yielding 12% performance gains on LIBERO and real-world tasks with 10-20% added samples.

  17. WorldVLA: Towards Autoregressive Action World Model

    cs.RO 2025-06 unverdicted novelty 5.0

    WorldVLA unifies VLA and world models in one autoregressive system, shows they boost each other, and adds an attention mask to stop error buildup when generating action chunks.