Metraipolis › archive › dc_7466ce95bc67
IMLE-VLA: Fast Single-Step Action Generation for Vision-Language-Action Policies — arXiv:2609.10915
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.10915
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T06:35:49.495Z
- SHA-256 of the archived text
6544fc73ad88448e79e58588fc32b4a9ddaf779d9c5ebd386a60d1c6ed8112d1- Stable URL of this page
https://metraipolis.com/documento/dc_7466ce95bc67
How to cite
IMLE-VLA: Fast Single-Step Action Generation for Vision-Language-Action Policies — arXiv:2609.10915 — arXiv (en), https://arxiv.org/abs/2609.10915 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7466ce95bc67 (Metraipolis archive dc_7466ce95bc67, sha256 6544fc73ad88448e79e58588fc32b4a9ddaf779d9c5ebd386a60d1c6ed8112d1)
BibTeX
@misc{dc7466ce95bc67,
title = {IMLE-VLA: Fast Single-Step Action Generation for Vision-Language-Action Policies — arXiv:2609.10915},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.10915},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7466ce95bc67; sha256 6544fc73ad88448e79e58588fc32b4a9ddaf779d9c5ebd386a60d1c6ed8112d1},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.10915. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Vision-language-action (VLA) policies leverage pretrained vision-language backbones to achieve strong cross-task generalization. A leading design couples this backbone with a dedicated continuous action head trained via diffusion or flow matching. However, such heads rely on iterative multi-step sampling, for example 10 Euler steps in $\pi_{0.5}$. This creates an inference bottleneck that produces stop-and-go movement in the robot and slower task completion. We introduce IMLE-VLA, which replaces the iterative action head with a single-step conditional generator trained via conditional Implicit Maximum Likelihood Estimation (cIMLE). The cIMLE objective promotes multimodal action coverage, avoiding the mode collapse of naive regression heads while eliminating multi-step sampling entirely. When IMLE-VLA is applied to $\pi_{0.5}$, it increases inference frequency 3.67x (55 Hz vs. 15 Hz), enabling up to 11x higher action throughput. On the 40-task LIBERO benchmark, IMLE-VLA achieves the highest average success rate (98.0%) among all baselines while leading in inference frequency. Under the test-time perturbations of LIBERO-plus, IMLE-VLA retains $\pi_{0.5}$'s robustness while other baselines degrade sharply, confirming that the cIMLE head preserves generalization. Real-world experiments on a Franka Emika Panda across four tasks demonstrate smoother motion (2.2x to 3.0x lower jerk) and faster task completion, with IMLE-VLA outperforming $\pi_{0.5}$ on every task and reducing average VLA inference time per episode by 3.9x to 6.6x. Videos and code are available at this https URL
Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-11 in cs.CV).
Categorie arXiv: Robotics (cs.RO) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.10915 — 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 2026-10-16T06:06:28.811Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).