Metraipolis › archive › dc_2f036fc56a5e
Q-WAM: 4-Bit Quantization of World Action Models with Action-Subspace Protection — arXiv:2609.33269
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.33269
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T23:55:06.818Z
- SHA-256 of the archived text
ad181aa9b01c28c00520aa076c26acb06bcbe3a266e32cf8f4f4ffc709728347- Stable URL of this page
https://metraipolis.com/documento/dc_2f036fc56a5e
How to cite
Q-WAM: 4-Bit Quantization of World Action Models with Action-Subspace Protection — arXiv:2609.33269 — arXiv (en), https://arxiv.org/abs/2609.33269 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2f036fc56a5e (Metraipolis archive dc_2f036fc56a5e, sha256 ad181aa9b01c28c00520aa076c26acb06bcbe3a266e32cf8f4f4ffc709728347)
BibTeX
@misc{dc2f036fc56a5e,
title = {Q-WAM: 4-Bit Quantization of World Action Models with Action-Subspace Protection — arXiv:2609.33269},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.33269},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2f036fc56a5e; sha256 ad181aa9b01c28c00520aa076c26acb06bcbe3a266e32cf8f4f4ffc709728347},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33269. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
World Action Models (WAMs) jointly generate video and robot actions through iterative diffusion and perform strongly in robotic manipulation. However, their prohibitive compute and memory costs pose substantial deployment challenges. Post-training quantization (PTQ) can reduce these costs, but existing PTQ methods such as smoothing and rotation are insufficient to maintain the precision of action generation. To overcome this limitation, we propose Q-WAM, a new 4-bit weight-activation quantization for WAMs that preserves the actions the model generates. Specifically, we introduce the \textit{Action Observability Gramian (AOG)}, which measures how much rounding errors in each weighted combination of a layer's input channels change the final action through all denoising steps. We also develop Action-Subspace Protection (ASP), which keeps the few most action-sensitive channel combinations in a tiny 16-bit low-rank branch and quantizes the complementary weights and activations to 4 bits, both as dense matrix multiplications that run efficiently on GPUs. Finally, to preserve action quality with minimal overhead, we identify the experts that matter most for the generated action by aggregating the AOG-derived action mass across the layers of each expert and apply ASP only to those experts. We evaluate Q-WAM on three WAMs, both in simulation and in real-world deployment. On the RoboTwin 2.0 benchmark, it reaches 89.6--93.0\% average success rate, within 1.1 percentage points of the 16-bit models, while reducing the memory of the quantized blocks by 3.1--3.4$\times$. Our method outperforms the strongest baseline, SVDQuant, by 2.5--8.7 percentage points. On a Unitree G1 humanoid and a bimanual UR3 robot, it improves success over SVDQuant by 12.8-17.6 percentage points.
Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CV).
Categorie arXiv: Robotics (cs.RO) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.33269 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-07T23:29:27.684Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).