Metraipolis › archive › dc_2cfa462cf35e

Alignment-Guided Flow Transformer for Efficient Vision-Language-Action Policy Learning — arXiv:2609.34467

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.34467
License (as archived): CC0-1.0
Archived: 2026-10-07T22:38:18.702Z
SHA-256 of the archived text: adcac0e273468a8d919fc1ee9d383c6fb606bbc4e8a497cd9de6bccea52fea3a
Stable URL of this page: https://metraipolis.com/documento/dc_2cfa462cf35e

How to cite

Alignment-Guided Flow Transformer for Efficient Vision-Language-Action Policy Learning — arXiv:2609.34467 — arXiv (en), https://arxiv.org/abs/2609.34467 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2cfa462cf35e (Metraipolis archive dc_2cfa462cf35e, sha256 adcac0e273468a8d919fc1ee9d383c6fb606bbc4e8a497cd9de6bccea52fea3a)

BibTeX

@misc{dc2cfa462cf35e,
  title = {Alignment-Guided Flow Transformer for Efficient Vision-Language-Action Policy Learning — arXiv:2609.34467},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.34467},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2cfa462cf35e; sha256 adcac0e273468a8d919fc1ee9d383c6fb606bbc4e8a497cd9de6bccea52fea3a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.34467. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Recent advances in Vision-Language-Action (VLA) models point toward general-purpose robotic intelligence by unifying perception, instruction, and control. Despite impressive progress, existing VLA models often adapt poorly due to \emph{tri-modal misalignment} among vision, language, and action, which weakens action grounding and hurts generalization and fine-tuning efficiency. In this work, we present Alignment-Guided Flow Transformer (AGFT), a novel framework that explicitly enforces tri-modal alignment through a dedicated alignment loss, bridging the representational gap across modalities and enhancing task adaptation. While prior research has predominantly emphasized bi-modal vision--language alignment, we systematically formalize and study tri-modal alignment in VLA models, and provide both ablations and analysis to isolate its role in improving adaptation and robustness. To further accelerate deployment, we adopt a flow-matching objective, enabling substantially fewer inference steps than diffusion-based policies while maintaining accuracy. Theoretically, we establish a quantitative connection between the tri-modal alignment gap and the optimization tightness of flow matching; empirically, experiments on the extensive benchmark show that AGFT achieves superior success rates and lower inference latency compared to SOTA baselines, underscoring tri-modal alignment as a key ingredient for scaling robust VLA manipulation.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.34467 — 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 2026-10-07T22:14:28.199Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).