aiqbench analysis
This paper addresses the "discretization bottleneck" in Vision-Language-Action models by introducing $\mathcal{M}^2$Tok, a tokenizer that uses multi-head decomposition and independent codebooks to better preserve fine-grained action dynamics. The approach significantly reduces reconstruction loss compared to existing methods, directly enhancing the policy performance of VLA models on both simulated and real-world robotic tasks. By expanding representational expressivity through combinatorial codebook mechanisms, this work removes a key barrier to scaling autoregressive models for precise robotic control.
${M}^2$Tok: Multi-head Multi-codebook Discrete Action Tokenization for Vision-Language-Action Models
arXiv:2609.18259v1 Announce Type: cross Abstract: Recent advancements have successfully adapted autoregressive language models to process multimodal signals, such as images and actions. Since raw action signals are continuous, effective tokenization is essential to map high-dimensional inputs into compact discrete tokens for autoregressive processing. However, existing discrete action tokenizers often suffer from high reconstruction loss, failing to preserve the fine-grained dynamics required for precise control. This ``discretization bottleneck'' significantly limits the performance ceiling of downstream Vision-Language-Action (VLA) models. To address this, we propose $\mathcal{M}^2$Tok, a Multi-head Multi-codebook Action Tokenizer designed to minimize reconstruction error and enhance policy performance. Our approach introduces two key structural innovations: (1) we decompose the latent action features into multiple heads, enabling the model to implicitly align specific heads with distinct action dimensions; (2) we assign independent codebooks to each head for quantization. By leveraging the combinatorial nature of multiple codebooks, we significantly expand the representational expressivity of the tokenizer, leading to substantially lower reconstruction loss compared to previous methods. We evaluate the $\mathcal{M}^2$Tok-based VLA on the RoboTwin, Simpler-Env, and 3 zero-shot real-world tasks. Experimental results demonstrate our method not only achieves superior reconstruction fidelity but also significantly boosts the success rate of VLA models. Comprehensive ablation studies further confirm the effectiveness of the multi-head and multi-codebook mechanisms. Code is available at \href{https://github.com/cpaaax/M2Tok}{https://github.com/cpaaax/M2Tok}.
Abstract:Recent advancements have successfully adapted autoregressive language models to process multimodal signals, such as images and actions. Since raw action signals are continuous, effective tokenization is essential to map high-dimensional inputs into compact discrete tokens for autoregressive processing. However, existing discrete action tokenizers often suffer from high reconstruction loss, failing to preserve the fine-grained dynamics required for precise control. This ``discretization bottleneck'' significantly limits the performance ceiling of downstream Vision-Language-Action (VLA) models. To address this, we propose $\mathcal{M}^2$Tok, a Multi-head Multi-codebook Action Tokenizer designed to minimize reconstruction error and enhance policy performance. Our approach introduces two key structural innovations: (1) we decompose the latent action features into multiple heads, enabling the model to implicitly align specific heads with distinct action dimensions; (2) we assign independent codebooks to each head for quantization. By leveraging the combinatorial nature of multiple codebooks, we significantly expand the representational expressivity of the tokenizer, leading to substantially lower reconstruction loss compared to previous methods. We evaluate the $\mathcal{M}^2$Tok-based VLA on the RoboTwin, Simpler-Env, and 3 zero-shot real-world tasks. Experimental results demonstrate our method not only achieves superior reconstruction fidelity but also significantly boosts the success rate of VLA models. Comprehensive ablation studies further confirm the effectiveness of the multi-head and multi-codebook mechanisms. Code is available at \href{this https URL}{this https URL}.
Robotics (cs.RO); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
arXiv:2609.18259 [cs.RO]
(or arXiv:2609.18259v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2609.18259 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 321 words
Related coverage
NAC: Neural Action Codec for Vision-Language-Action Models
arXiv · Sep 28, 2026
CrossVLA: Cross-Paradigm Post-Training and Inference Optimization for Vision-Language-Action Models
arXiv · Jun 9, 2026
Continuous Actions from Discrete Minds: Latent-Aligned Planning for End-to-End Autonomous Driving
arXiv · Sep 4, 2026
Suppression Sticks, Locality Is Fragile: A Closed-Loop Target-and-Control Audit of Task-Vector Negation in VLA Policies
arXiv · Aug 6, 2026