Add CITATION.cff

CITATION.cff

@@ -1,35 +1,18 @@
 cff-version: 1.2.0
-message: "If you use this benchmark or any of the trained models, please cite it as below."
-title: "Fundus 9-Model Benchmark: VGG19, ResNet50/101, DenseNet121, InceptionV3, Swin-B, CLIP, DINOv2-L, RETFound"
+message: "If you use this benchmark, please cite it as below."
+title: "A 9-Backbone Benchmark for Fundus Image Classification on a 10-Class Mendeley Dataset"
 authors:
-  - family-names: Katiyo
-    given-names: Daryl Panashe
-    affiliation: Xidian University
-version: "1.0.0"
-date-released: 2026-05-26
-license: Apache-2.0
+  - family-names: "Katiyo"
+    given-names: "Daryl Panashe"
 type: software
-repository-code: "https://huggingface.co/DoB24/fundus-9model-benchmark"
+date-released: 2026-05-26
 url: "https://huggingface.co/DoB24/fundus-9model-benchmark"
+repository-code: "https://huggingface.co/DoB24/fundus-9model-benchmark"
+license: Apache-2.0
 keywords:
   - fundus
   - retinal-imaging
-  - medical-image-classification
   - deep-learning
+  - medical-imaging
+  - classification
   - benchmark
-  - RETFound
-  - DINOv2
-  - vision-transformer
-abstract: >
-  A reproducible 10-class fundus disease classification benchmark with nine deep
-  models (VGG19, ResNet50, ResNet101, DenseNet121, InceptionV3, Swin-B, CLIP-ViT-B/16,
-  DINOv2-L, RETFound) trained under an identical v2 protocol (CLAHE + RandAugment +
-  WeightedRandomSampler + MixUp/CutMix + 6-view TTA) on a pHash-grouped 5-fold split
-  of the Mendeley 10-class fundus dataset.
-references:
-  - type: dataset
-    title: "Eye disease image dataset"
-    authors:
-      - family-names: Khaled
-        given-names: Mohammad
-    url: "https://data.mendeley.com/datasets/s9bfhswzjb/1"