mlsanityData to 7 Oct 2026

Co-VLA: Consensus-based Federated Training for Vision-Language-Action Models

arXiv 2609.19923 ↗First posted 17 Sep 2026cs.ROcited by 0

On the map: Vision-language-action models → Generalist VLA models and benchmarks also fits: Reinforcement learning and post-training

Abstract (arXiv)

Vision-language-action models (VLAs) have emerged as a promising paradigm for general-purpose robot learning, with performance improving as models and datasets scale. Scaling robot data collection, however, remains challenging because data are naturally distributed across robots, tasks, and locations, making centralization costly or impractical. Federated learning offers a way to train on decentralized robot data, but applying it to VLAs requires accounting for heterogeneous robot client data distributions. We present Co-VLA, which applies consensus optimization using the Alternating Direction Method of Multipliers~(ADMM) to federated VLA training. We show that the same algorithm supports both full-model training and parameter-efficient fine-tuning with both fixed-rank and rank-adaptive adapters. The name Co-VLA reflects both consensus and collaboration: clients with different local robot datasets collaboratively train a shared model without sharing their data. Our experiments demonstrate that Co-VLA achieves performance comparable to centralized training in both full-model training and parameter-efficient fine-tuning settings. The project website and videos of our real-world experiments are available at https://embodiedvision.github.io/co-vla/.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Generalist VLA models and benchmarks, one of 13 directions in Vision-language-action models. Builds on 31 of its 39 references that are in the map. First posted on arXiv 17 Sep 2026.

Builds on

31 of 39 references.

Closest ancestors · 6

Same direction, newest first

  1. 2025X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Model11 Oct
  2. 2025FedVLA: Federated Vision-Language-Action Learning with Dual Gating Mixture-of-Experts for Robotic Manipulation4 Aug
  3. 2025SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics2 Jun
  4. 2025Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success27 Feb
  5. 2024OpenVLA: An Open-Source Vision-Language-Action Model13 Jun
  6. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control28 Jul

Other references · 25

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  2. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
  3. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,007
  4. 2024Octo: An Open-Source Generalist Robot Policycited by 1,427
  5. 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,338
  6. 2019On the Convergence of FedAvg on Non-IID Datacited by 1,220
  7. 2020Adaptive Federated Optimizationcited by 1,036
  8. 2023BridgeData V2: A Dataset for Robot Learning at Scalecited by 634
  9. 2024Improving LoRA in Privacy-preserving Federated Learningcited by 158
  10. 2024FLoRA: Federated Fine-Tuning Large Language Models with Heterogeneous Low-Rank Adaptationscited by 144
  11. 2020FedPD: A Federated Learning Framework with Optimal Rates and Adaptivity to Non-IID Datacited by 143
  12. 2023DiLoCo: Distributed Low-Communication Training of Language Modelscited by 115
13 more
  1. 2024Heterogeneous LoRA for Federated Fine-tuning of On-Device Foundation Modelscited by 114
  2. 2023Sparse Low-rank Adaptation of Pre-trained Language Modelscited by 96
  3. 2024Federated Fine-tuning of Large Language Models under Heterogeneous Tasks and Client Resourcescited by 93
  4. 2023SLoRA: Federated Parameter Efficient Fine-Tuning of Language Modelscited by 92
  5. 2024Selective Aggregation for Low-Rank Adaptation in Federated Learningcited by 80
  6. 2024FMB: a Functional Manipulation Benchmark for Generalizable Robotic Learningcited by 79
  7. 2022Federated Learning via Inexact ADMMcited by 44
  8. 2023pFedLoRA: Model-Heterogeneous Personalized Federated Learning with LoRA Tuningcited by 38
  9. 2022FedADMM: A Federated Primal-Dual Algorithm Allowing Partial Participationcited by 35
  10. 2015A General Analysis of the Convergence of ADMMcited by 22
  11. 2024Personalized Federated Fine-Tuning for LLMs via Data-Driven Heterogeneous Model Architecturescited by 10
  12. 2024Distributed Event-Based Learning via ADMMcited by 3
  13. 2022Distributed optimization on directed graphs based on inexact ADMM with partial participationcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026An Empirical Study and Open Testbed for Federated Fine-Tuning of Vision-Language-Action Models
  2. 2026ForgeVLA: Federated Vision-Language-Action Learning without Language Annotations
  3. 2026Cooperative Multi-Agent Vision-Language-Action Models via Reinforced Fine Tuning
  4. 2026VLAFlow: A Unified Training Framework for Vision-Language-Action Models via Co-training and Future Latent Alignment
  5. 2026Generalizable VLA Finetuning via Representation Anchoring and Language-Action Alignment
  6. 2025Vision-Language-Action Models for Robotics: A Review Towards Real-World Applications
  7. 2025VLA-0: Building State-of-the-Art VLAs with Zero Modification
  8. 2026FineVLA: Fine-Grained Instruction Alignment for Steerable Vision-Language-Action Policies