Add paper link to POVBench dataset card

README.md

@@ -1,25 +1,25 @@
 ---
-pretty_name: POVBench
-license: cc-by-4.0
 language:
-  - en
-task_categories:
-  - visual-question-answering
-  - image-text-to-text
+- en
+license: cc-by-4.0
 size_categories:
-  - 1K<n<10K
+- 1K<n<10K
 source_datasets:
-  - original
+- original
+task_categories:
+- visual-question-answering
+- image-text-to-text
+pretty_name: POVBench
 tags:
-  - image
-  - contextual-observer-grounding
-  - spatial-reasoning
-  - perspective-taking
-  - vision-language-models
-  - embodied-ai
-  - benchmark
-  - ai2thor
-  - procthor
+- image
+- contextual-observer-grounding
+- spatial-reasoning
+- perspective-taking
+- vision-language-models
+- embodied-ai
+- benchmark
+- ai2thor
+- procthor
 ---
 
 # POVBench
@@ -27,7 +27,7 @@ tags:
 **Contextual Observer Grounding: Evaluating Situated Spatial Reasoning in Vision-Language Models**  
 EMNLP 2026 Findings
 
-[Project page](https://mimo-owl.github.io/POVBench/) · [Code](https://github.com/mimo-owl/povbench-eval)
+[Project page](https://mimo-owl.github.io/POVBench/) · [Code](https://github.com/mimo-owl/povbench-eval) · [Paper](https://huggingface.co/papers/2609.06880)
 
 ![POVBench overview](assets/overview.png)
 
@@ -171,4 +171,4 @@ for known caveats.
   booktitle = {Findings of the Association for Computational Linguistics: EMNLP 2026},
   year      = {2026}
 }
-```
+```