Datasets:
Modalities:
Image
Formats:
imagefolder
Languages:
English
Size:
1K - 10K
Tags:
image
contextual-observer-grounding
spatial-reasoning
perspective-taking
vision-language-models
embodied-ai
License:
Unify venue wording
Browse files
meta.json
CHANGED
|
@@ -4,7 +4,7 @@
|
|
| 4 |
"description": "A benchmark for contextual observer grounding: localizing a target object from an observer-relative natural-language sentence, under three levels of observer specification.",
|
| 5 |
"paper": {
|
| 6 |
"title": "Contextual Observer Grounding: Evaluating Situated Spatial Reasoning in Vision-Language Models",
|
| 7 |
-
"venue": "
|
| 8 |
"project_page": "https://mimo-owl.github.io/POVBench/",
|
| 9 |
"code": "https://github.com/mimo-owl/povbench-eval"
|
| 10 |
},
|
|
|
|
| 4 |
"description": "A benchmark for contextual observer grounding: localizing a target object from an observer-relative natural-language sentence, under three levels of observer specification.",
|
| 5 |
"paper": {
|
| 6 |
"title": "Contextual Observer Grounding: Evaluating Situated Spatial Reasoning in Vision-Language Models",
|
| 7 |
+
"venue": "EMNLP 2026 Findings",
|
| 8 |
"project_page": "https://mimo-owl.github.io/POVBench/",
|
| 9 |
"code": "https://github.com/mimo-owl/povbench-eval"
|
| 10 |
},
|