| --- |
| license: apache-2.0 |
| language: |
| - en |
| base_model: robbyant/lingbot-va-base |
| tags: |
| - robotics |
| - world-action-model |
| - event-aligned |
| - robotwin |
| - custom-code |
| --- |
| |
| # Event-Aligned — RoboTwin 2.0 |
|
|
| Event-Aligned model for RoboTwin 2.0 bimanual manipulation, built on pretrained [LingBot-VA base](https://huggingface.co/robbyant/lingbot-va-base). |
|
|
| [Code][code-link] | [Models][models-link] | [Project page][project-page-link] |
|
|
| <p align="center"> |
| <a href="assets/teaser.png"><img src="assets/teaser.png" alt="Visual-action sampling and DOMINO success rates under a matched training sequence budget: Episode Uniform 29.17%, Event-Aligned 42.83%." width="800"></a> |
| <br> |
| <em>Event-aligned visual-action sampling (left) and DOMINO success rates under a matched training sequence budget (right).</em> |
| </p> |
|
|
| ## Method |
|
|
| - **Event-aligned supervision:** preserve dense interaction evidence and sample transitions more sparsely, keeping training chunks within event boundaries. |
| - **Execution validity prediction:** identify valid action slots and skip redundant actions introduced by repeated-frame padding. |
|
|
| Inference uses observations and a task instruction. No event annotations are required during inference. |
|
|
| ## Download |
|
|
| Download the RoboTwin model from the [model collection][models-link]: |
|
|
| ```bash |
| python -m pip install -U huggingface_hub |
| hf download Xun49/Event-aligned-robotwin --local-dir ./model/event-aligned_robotwin |
| ``` |
|
|
| Download the complete repository, including `transformer/`, `vae/`, `text_encoder/`, and `tokenizer/`. See the [code repository][code-link] for setup, inference, and evaluation. |
|
|
| ## Results reported in the paper |
|
|
| | Method | Clean SR (%) | Randomized SR (%) | Average SR (%) | |
| | --- | ---: | ---: | ---: | |
| | Event-Aligned | **93.02** | **91.58** | **92.30** | |
|
|
| Evaluation covers 50 RoboTwin 2.0 tasks under clean and randomized conditions, with 100 episodes per task for each condition. |
|
|
| ## License and attribution |
|
|
| The model weights are released under the [Apache License 2.0](LICENSE.txt). |
|
|
| Built on the Robbyant Team's [LingBot-VA](https://github.com/Robbyant/lingbot-va/tree/7c6ffa9bfc4b83582cafc860fab4c82cc7deeeeb). The companion code is also Apache-2.0 licensed. Retain applicable upstream licenses and attribution. Please also credit *Causal World Modeling for Robot Control*. |
|
|
| ## Citation |
|
|
| ```bibtex |
| @misc{yang2026eventaligned, |
| title = {Event-Aligned Visual Action Reasoning for World Action Models}, |
| author = {Yang, Xiaomeng and Wu, Yushu and Gao, Yi and Lei, Yuhao and Zhang, Xuan and Zhao, Pu and Wang, Yanzhi}, |
| year = {2026}, |
| eprint = {2610.09427}, |
| archivePrefix = {arXiv}, |
| primaryClass = {cs.CV}, |
| url = {https://arxiv.org/abs/2610.09427} |
| } |
| ``` |
|
|
| [code-link]: https://github.com/Xiaomeng-Yang/Event-aligned-WAM |
| [models-link]: https://huggingface.co/collections/Xun49/event-aligned |
| [project-page-link]: https://xiaomeng-yang.github.io/Event-aligned-WAM/ |
|
|
