Update README.md
Browse files
README.md
CHANGED
|
@@ -51,9 +51,14 @@ Five tasks are currently formulated in the framework, requiring only ordinary RG
|
|
| 51 |
- 🔥 **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.
|
| 52 |
- 🔥 **Naturally Verifiable:** Intrinsic supervisory signals determined by pretext objectives are naturally verifiable, aligning Spatial-SSRL well with the RLVR paradigm.
|
| 53 |
<p style="text-align: center;">
|
| 54 |
-
<img src="assets/
|
| 55 |
</p>
|
| 56 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 57 |
|
| 58 |
## 🛠️ Usage
|
| 59 |
Here we provide a code snippet for you to start a simple trial of <strong>Spatial-SSRL-Qwen3VL-4B</strong> on your own device. You can download the model from 🤗<a href="https://huggingface.co/internlm/Spatial-SSRL-Qwen3VL-4B">Spatial-SSRL-Qwen3VL-4B Model</a > before your trial!
|
|
|
|
| 51 |
- 🔥 **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.
|
| 52 |
- 🔥 **Naturally Verifiable:** Intrinsic supervisory signals determined by pretext objectives are naturally verifiable, aligning Spatial-SSRL well with the RLVR paradigm.
|
| 53 |
<p style="text-align: center;">
|
| 54 |
+
<img src="assets/comparison_v2.png" alt="Teaser" width="100%">
|
| 55 |
</p>
|
| 56 |
|
| 57 |
+
## 📊 Results
|
| 58 |
+
We train Qwen3-VL-4B-Instruct with our Spatial-SSRL paradigm and the experimental results across seven spatial understanding benchmarks are shown below.
|
| 59 |
+
<p style="text-align: center;">
|
| 60 |
+
<img src="assets/exp_result_new3.png" alt="Pipeline" width="100%">
|
| 61 |
+
</p>
|
| 62 |
|
| 63 |
## 🛠️ Usage
|
| 64 |
Here we provide a code snippet for you to start a simple trial of <strong>Spatial-SSRL-Qwen3VL-4B</strong> on your own device. You can download the model from 🤗<a href="https://huggingface.co/internlm/Spatial-SSRL-Qwen3VL-4B">Spatial-SSRL-Qwen3VL-4B Model</a > before your trial!
|