internlm
/

Spatial-SSRL-7B

Image-Text-to-Text

sptial understanding

self-supervised learning

text-generation-inference

Model card Files Files and versions

yuhangzang commited on 18 days ago

Commit

f018eae

·

verified ·

1 Parent(s): cf00570

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -50,7 +50,7 @@ Five tasks are currently formulated in the framework, requiring only ordinary RG
 - 🔥 **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.
 <p style="text-align: center;">
-  <img src="assets/comparison_1029final.png" alt="Teaser" width="100%">
 </p>
 ## 📊 Results

 - 🔥 **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.
 <p style="text-align: center;">
+  <img src="assets/comparison_v2.png" alt="Teaser" width="100%">
 </p>
 ## 📊 Results