# Dense Policy: Bidirectional Autoregressive Learning of Actions

arXiv: https://arxiv.org/abs/2503.13217 · HTML: https://mlsanity.com/paper/2503.13217/

## Key facts

- Cited by 19 papers in the map (data through 14 Oct 2026).
- Builds on 43 of its 54 references that are in the map.
- First posted on arXiv 17 Mar 2025.

## Builds on (most cited first, 30 of 43)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Continuous control with deep reinforcement learning](https://mlsanity.com/paper/1509.02971/) (2015)
- [RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control](https://mlsanity.com/paper/2307.15818/) (2023)
- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [Diffusion Policy: Visuomotor Policy Learning via Action Diffusion](https://mlsanity.com/paper/2303.04137/) (2023)
- [RT-1: Robotics Transformer for Real-World Control at Scale](https://mlsanity.com/paper/2212.06817/) (2022)
- [Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware](https://mlsanity.com/paper/2304.13705/) (2023)
- [Group Normalization](https://mlsanity.com/paper/1803.08494/) (2018)
- [Octo: An Open-Source Generalist Robot Policy](https://mlsanity.com/paper/2405.12213/) (2024)
- [4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks](https://mlsanity.com/paper/1904.08755/) (2019)
- [Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning](https://mlsanity.com/paper/1910.10897/) (2019)
- [Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction](https://mlsanity.com/paper/2404.02905/) (2024)
- [SpanBERT: Improving Pre-training by Representing and Predicting Spans](https://mlsanity.com/paper/1907.10529/) (2019)
- [Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrations](https://mlsanity.com/paper/1709.10087/) (2017)
- [FAST: Efficient Action Tokenization for Vision-Language-Action Models](https://mlsanity.com/paper/2501.09747/) (2025)
- [Autoregressive Image Generation without Vector Quantization](https://mlsanity.com/paper/2406.11838/) (2024)
- [SAPIEN: A SimulAted Part-based Interactive ENvironment](https://mlsanity.com/paper/2003.08515/) (2020)
- [BC-Z: Zero-Shot Task Generalization with Robotic Imitation Learning](https://mlsanity.com/paper/2202.02005/) (2022)
- [Implicit Behavioral Cloning](https://mlsanity.com/paper/2109.00137/) (2021)
- [VIMA: General Robot Manipulation with Multimodal Prompts](https://mlsanity.com/paper/2210.03094/) (2022)
- [ERNIE 2.0: A Continual Pre-training Framework for Language Understanding](https://mlsanity.com/paper/1907.12412/) (2019)
- [RH20T: A Comprehensive Robotic Dataset for Learning Diverse Skills in One-Shot](https://mlsanity.com/paper/2307.00595/) (2023)
- [ManipLLM: Embodied Multimodal Large Language Model for Object-Centric Robotic Manipulation](https://mlsanity.com/paper/2312.16217/) (2023)

## Built on by (most cited first, 19 of 19)

- [OpenHelix: A Short Survey, Empirical Analysis, and Open-Source Dual-System VLA Model for Robotic Manipulation](https://mlsanity.com/paper/2505.03912/) (2025)
- [RMBench: Memory-Dependent Robotic Manipulation Benchmark with Insights into Policy Design](https://mlsanity.com/paper/2603.01229/) (2026)
- [World Guidance: World Modeling in Condition Space for Action Generation](https://mlsanity.com/paper/2602.22010/) (2026)
- [CLAP: Contrastive Latent Action Pretraining for Learning Vision-Language-Action Models from Human Videos](https://mlsanity.com/paper/2601.04061/) (2026)
- [From Spatial to Actions: Grounding Vision-Language-Action Model in Spatial Foundation Priors](https://mlsanity.com/paper/2510.17439/) (2025)
- [VITA: Vision-to-Action Flow Matching Policy](https://mlsanity.com/paper/2507.13231/) (2025)
- [H3DP: Triply-Hierarchical Diffusion Policy for Visuomotor Learning](https://mlsanity.com/paper/2505.07819/) (2025)
- [XR-1: Towards Versatile Vision-Language-Action Models via Learning Unified Vision-Motion Representations](https://mlsanity.com/paper/2511.02776/) (2025)
- [FreqPolicy: Frequency Autoregressive Visuomotor Policy with Continuous Tokens](https://mlsanity.com/paper/2506.01583/) (2025)
- [RoboRouter: Training-Free Policy Routing for Robotic Manipulation](https://mlsanity.com/paper/2603.07892/) (2026)
- [Fast Visuomotor Policy for Robotic Manipulation](https://mlsanity.com/paper/2510.12483/) (2025)
- [Towards Generalist Game Players: An Investigation of Foundation Models in the Game Multiverse](https://mlsanity.com/paper/2605.09965/) (2026)
- [GauDP: Reinventing Multi-Agent Collaboration through Gaussian-Image Synergy in Diffusion Policies](https://mlsanity.com/paper/2511.00998/) (2025)
- [L1 Sample Flow for Efficient Visuomotor Learning](https://mlsanity.com/paper/2511.17898/) (2025)
- [CUBic: Coordinated Unified Bimanual Perception and Control Framework](https://mlsanity.com/paper/2605.13452/) (2026)
- [Referring-Aware Visuomotor Policy Learning for Closed-Loop Manipulation](https://mlsanity.com/paper/2604.05544/) (2026)
- [FocalPolicy: Frequency-Optimized Chunking and Locally Anchored Flow Matching for Coherent Visuomotor Policy](https://mlsanity.com/paper/2605.15944/) (2026)
- [Patch Policy: Efficient Embodied Control via Dense Visual Representations](https://mlsanity.com/paper/2607.18236/) (2026)
- [ActionUNet: Improving Robustness of VLA Models with Efficient Multi-scale Fine-tuning](https://mlsanity.com/paper/2609.34982/) (2026)

## Abstract (arXiv)

Mainstream visuomotor policies predominantly rely on generative models for holistic action prediction, while current autoregressive policies, predicting the next token or chunk, have shown suboptimal results. This motivates a search for more effective learning methods to unleash the potential of autoregressive policies for robotic manipulation. This paper introduces a bidirectionally expanded learning approach, termed Dense Policy, to establish a new paradigm for autoregressive policies in action prediction. It employs a lightweight encoder-only architecture to iteratively unfold the action sequence from an initial single frame into the target sequence in a coarse-to-fine manner with logarithmic-time inference. Extensive experiments validate that our dense policy has superior autoregressive learning capabilities and can surpass existing holistic generative policies. Our policy, example data, and training code will be publicly available upon publication. Project page: https: //selen-suyue.github.io/DspNet/.