Update README.md

README.md

@@ -115,46 +115,34 @@ This AI model can be embedded as an Application Programming Interface (API) call
 
 # Training and Evaluation Datasets
 
+
 ## Training Dataset
 
-** Data Modality
+**NV-CC-Img-Text-Dataset**
 
-NV-CC-Img-Text-Dataset  <br>
-** Data Modality <br>
-* Image <br>
-** Image Training Data Size <br>
-* 1 Million to 1 Billion Images <br>
-** Data Collection Method by dataset <br>
-* Automated <br>
-** Labeling Method by dataset <br>
-* Not Applicable (no labels are needed; supervision comes from teacher models via multi-teacher distillation) <br>
-**Properties:** ~172M total training samples processed over 300k optimizer steps (less than one epoch over the source dataset). Global batch size of 512 low-resolution images (sampled from 128, 192, 224, 256, 384, 432 px) plus 64 high-resolution images (from 512, 768, 1024, 1152 px). <br>
+* **Data Modality:** Image
+* **Image Training Data Size:** 1 Million to 1 Billion Images
+* **Data Collection Method by dataset:** Automated
+* **Labeling Method by dataset:** Not Applicable (no labels are needed; supervision comes from teacher models via multi-teacher distillation)
+* **Properties:** ~172M total training samples processed over 300k optimizer steps (less than one epoch over the source dataset). Global batch size of 512 low-resolution images (sampled from 128, 192, 224, 256, 384, 432 px) plus 64 high-resolution images (from 512, 768, 1024, 1152 px).
 
 ## Evaluation Datasets
 
-ADE20K <br>
-** Link <br>
-* [ADE20K](https://ade20k.csail.mit.edu/) <br>
-** Data Collection <br>
-* Manually-Collected <br>
-** Labeling Method <br>
-* Manually-Collected <br>
-** Training Images <br>
-* 25,574 <br>
-** Validation Images <br>
-* 2,000 <br>
-
-ImageNet <br>
-** Link <br>
-* [ImageNet](https://www.image-net.org/) <br>
-** Data Collection <br>
-* Automated <br>
-** Labeling Method <br>
-* Manually-Collected <br>
-** Training Images <br>
-* 1,281,167 <br>
-** Validation Images <br>
-* 50,000 <br>
+**ADE20K**
+
+* **Link:** [ADE20K](https://ade20k.csail.mit.edu/)
+* **Data Collection:** Manually-Collected
+* **Labeling Method:** Manually-Collected
+* **Training Images:** 25,574
+* **Validation Images:** 2,000
+
+**ImageNet**
+
+* **Link:** [ImageNet](https://www.image-net.org/)
+* **Data Collection:** Automated
+* **Labeling Method:** Manually-Collected
+* **Training Images:** 1,281,167
+* **Validation Images:** 50,000
 
 For downstream VLM evaluation, RADIO1D was paired with the Nemotron-Nano-9B-v2 LLM in the Nemotron VL framework and evaluated on TextVQA, DocVQA, InfoVQA, OCRBench, OCRBench v2 (EN/CN), AI2D, ChartQA, MMMU, SeedBench, and LongVideoBench.