EUVIMEDEUVIMED entdecken
Crossref

DiTEA: Mixture-of-Experts for Vision-Language-Action Model in Robotic Manipulation

Chengxuan Li, Xingwan Wang

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (22) · 18379-18387

DOI: 10.1609/aaai.v40i22.38902 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Chengxuan Li und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The current diffusion-based Vision-Language-Action (VLA) models have faster inference speed and the ability to solve the action muti-modality problem in robot manipulation tasks compared to traditional autoregressive models after large-scale pre-training and post-training. However, the diffusion-based VLA models were found to have poor instruction-following ability, and after fine-tuning training on multiple tasks, them often suffer from "skill forgetting" due to conflicting model weights on each task. To address this problem, we propose DiTEA, a Diffusion Transformer-based Mixture-of-Experts (MoE) VLA model. Specifically, it fuses the MoE module into the action head of VLA to form Action MoE, and in addition, we design the Task-Instruction Gate, which uses language instructions to select specific experts for tasks they specialize in, in order to improve the VLA's instruction-following ability. We conducted comprehensive experiments and ablation study to evaluate the efficacy of our model under different designs. Experimental results from simulation and real-world show that our DiTEA has excellent improvement in multi-task compared to baseline and other VLAs.

Zitieren

APA

Li, C., & Wang, X. (2026). DiTEA: Mixture-of-Experts for Vision-Language-Action Model in Robotic Manipulation. Proceedings of the AAAI Conference on Artificial Intelligence, 40(22), 18379-18387. https://doi.org/10.1609/aaai.v40i22.38902

Vancouver

Li C, Wang X. DiTEA: Mixture-of-Experts for Vision-Language-Action Model in Robotic Manipulation. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(22):18379-18387. doi:10.1609/aaai.v40i22.38902

BibTeX
@article{Li2026,
  author = {Li, Chengxuan and Wang, Xingwan},
  title = {DiTEA: Mixture-of-Experts for Vision-Language-Action Model in Robotic Manipulation},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {22},
  pages = {18379--18387},
  doi = {10.1609/aaai.v40i22.38902}
}
Kostenlos starten →