# MergeVLA: Cross-Skill Model Merging Toward a Generalist Vision-Language-Action Agent

arXiv: https://arxiv.org/abs/2511.18810 · HTML: https://mlsanity.com/paper/2511.18810/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- Placed in Embodied and 3D vision-language models, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 64 of its 69 references that are in the map.
- First posted on arXiv 24 Nov 2025.

## Placed in

- Multimodal LLMs / vision-language models / [Embodied and 3D vision-language models](https://mlsanity.com/mllm/embodied-and-3d-vision-language-models/)
- Vision-language-action models / [Generalist VLA models and benchmarks](https://mlsanity.com/vla/generalist-vla-models-and-benchmarks/)
- LLM agents and tool use / [Multimodal and embodied agents](https://mlsanity.com/agents/multimodal-and-embodied-agents/)

## Builds on (most cited first, 30 of 64)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control](https://mlsanity.com/paper/2307.15818/) (2023)
- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024)
- [RT-1: Robotics Transformer for Real-World Control at Scale](https://mlsanity.com/paper/2212.06817/) (2022)
- [π0.5: a Vision-Language-Action Model with Open-World Generalization](https://mlsanity.com/paper/2504.16054/) (2025)
- [Octo: An Open-Source Generalist Robot Policy](https://mlsanity.com/paper/2405.12213/) (2024)
- [LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning](https://mlsanity.com/paper/2306.03310/) (2023)
- [Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time](https://mlsanity.com/paper/2203.05482/) (2022)
- [GR00T N1: An Open Foundation Model for Generalist Humanoid Robots](https://mlsanity.com/paper/2503.14734/) (2025)
- [Editing Models with Task Arithmetic](https://mlsanity.com/paper/2212.04089/) (2022)
- [Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success](https://mlsanity.com/paper/2502.19645/) (2025)
- [TIES-Merging: Resolving Interference When Merging Models](https://mlsanity.com/paper/2306.01708/) (2023)
- [QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation](https://mlsanity.com/paper/1806.10293/) (2018)
- [Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch](https://mlsanity.com/paper/2311.03099/) (2023)
- [BridgeData V2: A Dataset for Robot Learning at Scale](https://mlsanity.com/paper/2308.12952/) (2023)
- [Linear Mode Connectivity and the Lottery Ticket Hypothesis](https://mlsanity.com/paper/1912.05671/) (2019)
- [Merging Models with Fisher-Weighted Averaging](https://mlsanity.com/paper/2111.09832/) (2021)
- [CALVIN: A Benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Tasks](https://mlsanity.com/paper/2112.03227/) (2021)
- [SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics](https://mlsanity.com/paper/2506.01844/) (2025)
- [SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Model](https://mlsanity.com/paper/2501.15830/) (2025)
- [RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation](https://mlsanity.com/paper/2506.18088/) (2025)
- [Dataless Knowledge Fusion by Merging Weights of Language Models](https://mlsanity.com/paper/2212.09849/) (2022)
- [What is being transferred in transfer learning?](https://mlsanity.com/paper/2008.11687/) (2020)
- [Prismatic VLMs: Investigating the Design Space of Visually-Conditioned Language Models](https://mlsanity.com/paper/2402.07865/) (2024)
- [TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulation](https://mlsanity.com/paper/2409.12514/) (2024)
- [LIBERO-Plus: In-depth Robustness Analysis of Vision-Language-Action Models](https://mlsanity.com/paper/2510.13626/) (2025)
- [Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities](https://mlsanity.com/paper/2408.07666/) (2024)
- [Bridge Data: Boosting Generalization of Robotic Skills with Cross-Domain Datasets](https://mlsanity.com/paper/2109.13396/) (2021)
- [MT-Opt: Continuous Multi-Task Robotic Reinforcement Learning at Scale](https://mlsanity.com/paper/2104.08212/) (2021)

## Built on by (most cited first, 18 of 18)

- [Towards Long-Lived Robots: Continual Learning VLA Models via Reinforcement Fine-Tuning](https://mlsanity.com/paper/2602.10503/) (2026)
- [VLANeXt: Recipes for Building Strong VLA Models](https://mlsanity.com/paper/2602.18532/) (2026)
- [CORAL: Scalable Multi-Task Robot Learning via LoRA Experts](https://mlsanity.com/paper/2603.09298/) (2026)
- [Mimic Intent, Not Just Trajectories](https://mlsanity.com/paper/2602.08602/) (2026)
- [Generative Recall, Dense Reranking: Learning Multi-View Semantic IDs for Efficient Text-to-Video Retrieval](https://mlsanity.com/paper/2601.21193/) (2026)
- [Learning Action Manifold with Multi-view Latent Priors for Robotic Manipulation](https://mlsanity.com/paper/2605.11832/) (2026)
- [VLA-Pro: Cross-Task Procedural Memory Transfer for Vision-Language-Action Models](https://mlsanity.com/paper/2605.29562/) (2026)
- [LASER: A Corrective Lens for LVLMs via Visual Attention Preservation and Sink Suppression](https://mlsanity.com/paper/2607.01707/) (2026)
- [Suppression Sticks, Locality Is Fragile: A Closed-Loop Target-and-Control Audit of Task-Vector Negation in VLA Policies](https://mlsanity.com/paper/2608.04692/) (2026)
- [CapVector: Learning Transferable Capability Vectors in Parametric Space for Vision-Language-Action Models](https://mlsanity.com/paper/2605.10903/) (2026)
- [Domain Arithmetic: One-Shot VLA Adaptation under Environmental Shifts](https://mlsanity.com/paper/2607.00666/) (2026)
- [Fine-Tuning VLAs with Self-Demonstrated Generative Control for Multi-Task Manipulation](https://mlsanity.com/paper/2608.19490/) (2026)
- [Routing Drift Alone Does Not Diagnose Failure in Merged MoE LLMs](https://mlsanity.com/paper/2609.32821/) (2026)
- [Train Together or Merge Later? Unifying VLA Experts via a Shared Action Interface](https://mlsanity.com/paper/2609.33125/) (2026)
- [Does Adversarial Training Improve Generalization in Multi-View VLAs? Revealing and Mitigating View Collapse](https://mlsanity.com/paper/2609.33707/) (2026)
- [Triggering Generalist Reasoning via Predictive Uncertainty for Dual-System VLA](https://mlsanity.com/paper/2610.05025/) (2026)
- [SkillFusion: Towards Reliable Multi-Skill Composition for LLM Agents](https://mlsanity.com/paper/2610.15481/) (2026)
- [TacitVLA: Reasoning Vision-Language-Action Models Without Reasoning Latency](https://mlsanity.com/paper/2610.17540/) (2026)

## Abstract (arXiv)

Recent Vision-Language-Action (VLA) models reformulate vision-language models by tuning them with millions of robotic demonstrations. While they perform well when fine-tuned for a single embodiment or task family, extending them to multi-skill settings remains challenging: directly merging VLA experts trained on different tasks results in near-zero success rates. This raises a fundamental question: what prevents VLAs from mastering multiple skills within one model? With an empirical decomposition of learnable parameters during VLA fine-tuning, we identify two key sources of non-mergeability: (1) Finetuning drives LoRA adapters in the VLM backbone toward divergent, task-specific directions beyond the capacity of existing merging methods to unify. (2) Action experts develop inter-block dependencies through self-attention feedback, causing task information to spread across layers and preventing modular recombination. To address these challenges, we present MergeVLA, a merging-oriented VLA architecture that preserves mergeability by design. MergeVLA introduces sparsely activated LoRA adapters via task masks to retain consistent parameters and reduce irreconcilable conflicts in the VLM. Its action expert replaces self-attention with cross-attention-only blocks to keep specialization localized and composable. When the task is unknown, it uses a test-time task router to adaptively select the appropriate task mask and expert head from the initial observation, enabling unsupervised task inference. Across LIBERO, LIBERO-Plus, RoboTwin, and multi-task experiments on the real SO101 robotic arm, MergeVLA achieves performance comparable to or even exceeding individually finetuned experts, demonstrating robust generalization across tasks, embodiments, and environments. Project page: https://mergevla.github.io/