Update README.md

README.md

@@ -14,12 +14,16 @@ should probably proofread and complete it, then remove this comment. -->
 
 # trojblue/distill-q-align-aesthetic-siglip2-base
 
-This model is a fine-tuned version of [google/siglip2-base-patch16-512](https://huggingface.co/google/siglip2-base-patch16-512) on an unknown dataset.
-It achieves the following results on the evaluation set:
+This model is a fine-tuned version of [google/siglip2-base-patch16-512](https://huggingface.co/google/siglip2-base-patch16-512), specialized for predicting image aesthetics in anime illustrations based on Q-Align's quality metrics.
+
+It achieves the following performance on the evaluation set:
 - Loss: 0.0225
 - **Mse: 0.0225**
 - Rmse: 0.1502
 
+**NOTE: the Q-align model was not well-suited for anime aesthetics**,  and more or less have a "western" feel to it. The distilled version of Q-Align *quality* score may be slightly preferrable for actual usages.
+- (See: [trojblue/distill-q-align-quality-siglip2-base](https://huggingface.co/trojblue/distill-q-align-quality-siglip2-base)) 
+
 ## Model Description
 
 This model is a distilled version of the original [Q-Align](https://github.com/Q-Future/Q-Align) model, using Siglip2-base as its backbone. It is designed for simpler, faster, and more practical deployment compared to the original model. Benefits include: