InternLM 开源 Spatial-SSRL：用自监督强化学习提升空间理解
GitHub - InternLM/Spatial-SSRL: [CVPR 2026] Official release of "Spatial-SSRL: Enhancing Spatial Understanding via Self-Supervised Reinforcement Learning" · GitHub
InternLM 发布 Spatial-SSRL，一种自监督强化学习范式，用普通 RGB 与 RGB-D 图像替代人工标注数据来增强 LVLM 空间理解。该工作已被 CVPR 2026 接收，并开源 Spatial-SSRL-7B、Spatial-SSRL-81k 数据集，以及基于 Qwen2.5-VL-3B-Instruct 的 Spatial-SSRL-3B 和基于 Qwen3-VL-4B-Instruct 的 Spatial-SSRL-Qwen3VL-4B 模型。训练代码已于 2026/04/05 发布。
正文
当前显示原文，中文全文译文尚未就绪。
InternLM; Spatial-SSRL
Repository navigation
NotificationsYou must be signed in to change notification settings; Fork5 (5); Star142 (142)You must be signed in to star a repository
Repository files navigation
README; Apache-2.0 license
†Corresponding authors.
🚀 [2026/04/05] We have released the training code of Spatial-SSRL; 🚀 [2026/02/25] We have released the 🤗Spatial-SSRL-3B Model, initialized from Qwen2.5-VL-3B-Instruct; 🚀 [2026/02/21] Our work has been accepted by CVPR 2026; 🚀 [2025/11/24] We have released the 🤗Spatial-SSRL-Qwen3VL-4B Model, initialized from Qwen3-VL-4B-Instruct; 🚀 [2025/11/03] Now you can try out Spatial-SSRL-7B on 🤗Spatial-SSRL Space; 🚀 [2025/11/03] We have released the Spatial-SSRL-7B Model, and Spatial-SSRL-81k Dataset; 🚀 [2025/11/02] We have released the Spatial-SSRL repository
🌈 Overview
We are thrilled to introduce Spatial-SSRL, a novel self-supervised RL paradigm aimed at enhancing LVLM spatial understanding. By optimizing Qwen2.5-VL-7B with Spatial-SSRL, the model exhibits stronger spatial intelligence across seven spatial understanding benchmarks in both image and video settings.
💡 Highlights
🔥 Highly Scalable: Spatial-SSRL uses ordinary raw RGB and RGB-D images instead of richly-annotated public datasets or manual labels for data curation, making it highly scalable; 🔥 Cost-effective: Avoiding the need for human labels or API calls for general LVLMs throughout the entire pipeline endows Spatial-SSRL with cost-effectiveness; 🔥 Lightweight: Prior approaches for spatial understanding heavily rely on annotation of external tools, incurring inherent errors in training data and additional cost. In constrast, Spatial-SSRL is completely tool-free and can easily be extended to more self-supervised tasks; 🔥 Naturally Verifiable: Intrinsic supervisory signals determined by pretext objectives are naturally verifiable, aligning Spatial-SSRL well with the RLVR paradigm
📊 Results
We train Qwen2.5-VL-3B and Qwen2.5-VL-7B with our Spatial-SSRL paradigm and the experimental results across seven spatial understanding benchmarks are shown below.
⭐️ Quick Start
To directly experience Spatial-SSRL-7B, you can try it out on 🤗Spatial-SSRL Space!
Here we provide a code snippet for you to start a simple trial of Spatial-SSRL-7B on your own device. You can download the model from 🤗Spatial-SSRL-7B Model before your trial!
Here we provide a code snippet for you to start a simple trial of Spatial-SSRL-Qwen3VL-4B on your own device. You can download the model from 🤗Spatial-SSRL-Qwen3VL-4B Model before your trial!
Start your evaluation by referring to the tutorials in Eval.md
Cases
✒️Citation
If you find this project useful, please kindly cite:
📄 License
Usage and License Notices: The data and code are intended and licensed for research use only.
❤️ Acknowledgement
We extend our sincere gratitude to VLMEvalkit, the powerful toolkit to evaluate a vast range of LMMs!
The work is built upon EasyR1 and LlaMaFactory, two excellent codebase for LLM training!