Updated attributions

app.py

@@ -583,7 +583,7 @@ with gr.Blocks(title="Search-TTA (Simplified)", theme=gr.themes.Base()) as demo:
     # Footer to point out to model and data from app page.
     gr.Markdown(
         """
-        The satellite image CLIP encoder is fine-tuned using [Sentinel-2 Level 2A](https://docs.sentinel-hub.com/api/latest/data/sentinel-2-l2a/) satellite image and taxonomy images (with GPS locations) from [iNaturalist](https://inaturalist.org/). The sound CLIP encoder is fine-tuned with a subset of the same taxonomy images and their corresponding sounds from [iNaturalist](https://inaturalist.org/). These iNaturalist data are also used in [Taxabind](https://arxiv.org/abs/2411.00683). Note that while some of the examples above result in poor probability distributions, they will be improved using our test-time adaptation framework during the search process.
+        The satellite image CLIP encoder is fine-tuned using [Sentinel-2 Level 2A](https://docs.sentinel-hub.com/api/latest/data/sentinel-2-l2a/) satellite image and taxonomy images (with GPS locations) from [iNaturalist](https://inaturalist.org/). The sound CLIP encoder is fine-tuned with a subset of the same taxonomy images and their corresponding sounds from [iNaturalist](https://inaturalist.org/). Some of these iNaturalist data are also used in [Taxabind](https://arxiv.org/abs/2411.00683). Note that while some of the examples above result in poor probability distributions, they will be improved using our test-time adaptation framework during the search process.
         """
     )
 

app_multimodal_inference.py

@@ -374,7 +374,7 @@ with gr.Blocks(title="Search-TTA", theme=gr.themes.Base()) as demo:
     # Footer to point out to model and data from app page.
     gr.Markdown(
         """
-        The satellite image CLIP encoder is fine-tuned using [Sentinel-2 Level 2A](https://docs.sentinel-hub.com/api/latest/data/sentinel-2-l2a/) satellite image and taxonomy images (with GPS locations) from [iNaturalist](https://inaturalist.org/). The sound CLIP encoder is fine-tuned with a subset of the same taxonomy images and their corresponding sounds from [iNaturalist](https://inaturalist.org/). These iNaturalist data are also used in [Taxabind](https://arxiv.org/abs/2411.00683). Note that while some of the examples above result in poor probability distributions, they will be improved using our test-time adaptation framework during the search process.
+        The satellite image CLIP encoder is fine-tuned using [Sentinel-2 Level 2A](https://docs.sentinel-hub.com/api/latest/data/sentinel-2-l2a/) satellite image and taxonomy images (with GPS locations) from [iNaturalist](https://inaturalist.org/). The sound CLIP encoder is fine-tuned with a subset of the same taxonomy images and their corresponding sounds from [iNaturalist](https://inaturalist.org/). Some of these iNaturalist data are also used in [Taxabind](https://arxiv.org/abs/2411.00683). Note that while some of the examples above result in poor probability distributions, they will be improved using our test-time adaptation framework during the search process.
         """
     )