الملخص
Autoregressive Vision-Language-Action (VLA) models offer a scalable path to robot learning, yet existing action tokenizers treat tokenization as a compression problem, producing representations that are semantically misaligned with the autoregressive backbone. We propose CATok, a causal action tokenizer that reframes tokenization as a causally structured generative process. CATok introduces a conditional annealing mechanism that extracts action tokens by progressively annealing a flow-matching process: each token is conditioned on all preceding tokens and encodes the residual reconstruction signal at a specific noise level, establishing a coarse-to-fine causal token space whose generative semantics are structurally aligned with autoregressive modeling. A token-conditioned flow-matching decoder built on Multimodal Diffusion Transformer (MMDiT) reconstructs continuous action chunks from these discrete tokens with the precision of hybrid diffusion-head architectures. This discrete bottleneck enforces knowledge insulation by design, cleanly separating high-level semantic reasoning from low-level motor execution without requiring explicit attention masking. Extensive evaluations across three simulation benchmarks and real-world robotic manipulation tasks demonstrate that CATok consistently surpasses existing tokenization methods in both reconstruction fidelity-compression tradeoff and inference efficiency, while improving VLA task success rate and training efficiency, establishing a high-performance, scalable foundation for purely autoregressive VLA systems.
الكلمات المفتاحية
الموضوع
بيانات النشر
- المجلة
- غير متاح
- وصول مفتوح
- وصول مفتوح أخضر
اقتبس هذه المقالة
APA 7
Zhang, C., Cao, Y., Du, D., Lu, Y., Shao, J., Chen, R., Liu, J., Cao, L., Liu, Y., Zhao, H., & Xu, M. (2026). Rethinking Causal Action Tokenization with Conditional Annealing in Flow Matching. https://omanscience.com/ar/articles/rethinking-causal-action-tokenization-with-conditional-annealing-in-flow-matching
MLA 9
Zhang, Chenyu, et al. "Rethinking Causal Action Tokenization with Conditional Annealing in Flow Matching." https://omanscience.com/ar/articles/rethinking-causal-action-tokenization-with-conditional-annealing-in-flow-matching.
شيكاغو (المؤلف–التاريخ)
Zhang, Chenyu, Yuhang Cao, Daru Du, Yingxi Lu, Jing Shao, Ruoqu Chen, Jiajun Liu, Liu Cao, Yicheng Liu, Hang Zhao, and Mengdi Xu. 2026. "Rethinking Causal Action Tokenization with Conditional Annealing in Flow Matching." https://omanscience.com/ar/articles/rethinking-causal-action-tokenization-with-conditional-annealing-in-flow-matching.
هارفارد
Zhang, C., Cao, Y., Du, D., Lu, Y., Shao, J., Chen, R., Liu, J., Cao, L., Liu, Y., Zhao, H. and Xu, M. (2026) 'Rethinking Causal Action Tokenization with Conditional Annealing in Flow Matching', Available at: https://omanscience.com/ar/articles/rethinking-causal-action-tokenization-with-conditional-annealing-in-flow-matching.
فانكوفر
Zhang C, Cao Y, Du D, Lu Y, Shao J, Chen R, et al. Rethinking Causal Action Tokenization with Conditional Annealing in Flow Matching. https://omanscience.com/ar/articles/rethinking-causal-action-tokenization-with-conditional-annealing-in-flow-matching
IEEE
C. Zhang, Y. Cao, D. Du, Y. Lu, J. Shao, R. Chen, J. Liu, L. Cao, Y. Liu, H. Zhao, and M. Xu, "Rethinking Causal Action Tokenization with Conditional Annealing in Flow Matching," https://omanscience.com/ar/articles/rethinking-causal-action-tokenization-with-conditional-annealing-in-flow-matching.