CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulation
By Qixiu Li, Yaobo Liang, Zeyu Wang, Lin Luo, Xi Chen, M. Liao, Fangyun Wei, Yu Deng, Sicheng Xu, Yizhong Zhang and 8 more. arXiv.org, 2024
387 citations recorded by the source corpus.
The advancement of large Vision-Language-Action (VLA) models has significantly improved robotic manipulation in terms of language-guided task execution and generalization to unseen scenarios. While existing VLAs adapted from pretrained large Vision-Language-Models (VLM) have demonstrated promising generalizability, their task performance is still unsatisfactory as indicated by the low tasks success rates in different environments. In this paper, we present a new advanced VLA architecture derived from VLM. Unlike previous works that directly repurpose VLM for action prediction by simple action quantization, we propose a omponentized VLA architecture that has a specialized action module conditioned on VLM output. We systematically study the design of the action module and demonstrates the strong performance enhancement with diffusion action transformers for action sequence modeling, as well as their favorable scaling behaviors. We also conduct comprehensive experiments and ablation studies to evaluate the efficacy of our models with varied designs. The evaluation on 5 robot embodiments in simulation and real work shows that our model not only significantly surpasses existing VLAs in task performance and but also exhibits remarkable adaptation to new robots and generalization to unseen objects and backgrounds. It exceeds the average success rates of OpenVLA which has similar model size (7B) with ours by over 35% in simulated evaluation and 55% in real robot experiments. It also outperforms the large RT-2-X model (55B) by 18% absolute success rates in simulation. Code and models can be found on our project page (https://cogact.github.io/).
Subfields: Control and Systems Engineering
View Paper: https://www.semanticscholar.org/paper/54caf5927a6d656bb84d2a16e16fa2b0c5d6f522
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Control and Systems Engineering
- Sergey Levine
- Chelsea Finn
- Suraj Nair
- Karol Hausman
- Brian Ichter
- Karl Pertsch
- Pieter Abbeel
- Dibya Ghosh
Citations in the corpus
- SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Model (2025, neutral): “▷ Autoregressive paradigm with discretized action token has been widely adopted in existing VLAs [7, 30, 31], demonstrating significant advantages in large-scale training, seam-less LLM integration, and instruction-following capabilities.”
- RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation (2025, positive): “Vision–language–action (VLA) frameworks like OpenVLA [19] and CogACT [21], together with adaptations like Octo [34], LAPA [41], and OpenVLA-OFT [18] demonstrate efficient fine-tuning to novel robots and sensor modalities.”
- π0.5: a Vision-Language-Action Model with Open-World Generalization (2025, positive): “Vision-language-action models (VLAs) [23, 92, 42, 8, 83, 90, 55, 45, 3, 75, 64, 76, 84, 7, 37] offer an appealing solution: by fine-tuning pre-trained vision-language models for robot control, VLAs can leverage the semantic knowledge acquired from web-scale pretraining and bring it to bear on the…”
- A Survey on Vision--Language--Action Models for Embodied AI (2024, neutral): “SMART [65] presents a pretraining scheme encompassing three distinct objectives: forward dynamics prediction, inverse dynamics prediction, and randomly masked hindsight control.”