Update README.md

README.md

@@ -32,7 +32,7 @@ The MAEST models are also available for inference in the [Essentia](https://esse
 <!-- Provide the basic links for the model. -->
 
 - **Repository:** [MAEST](https://github.com/palonso/MAEST)
-- **Paper:** [Efficient Supervised Training of Audio Transformers for Music Representation Learning](http://hdl.handle.net/10230/58023)
+- **Paper:** [Efficient Supervised Training of Audio Transformers for Music Representation Learning](https://arxiv.org/abs/2309.16418)
 
 ## Uses
 
@@ -102,7 +102,7 @@ Our models were trained using Discogs20, [MTG](https://www.upf.edu/web/mtg) in-h
 
 <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
 
-Most training details are detailed in the [paper](http://hdl.handle.net/10230/58023) and [official implementation](https://github.com/palonso/MAEST/) of the model.
+Most training details are detailed in the [paper](https://arxiv.org/abs/2309.16418) and [official implementation](https://github.com/palonso/MAEST/) of the model.
 
 #### Preprocessing
 
@@ -112,7 +112,7 @@ In Transformers, this mel-spectrogram signature is replicated to a certain exten
 ## Evaluation, Metrics, and results
 
 The MAEST models were pre-trained in the task of music style classification, and their internal representations were evaluated via downstream MLP probes in several benchmark music understanding tasks.
-Check the original [paper](http://hdl.handle.net/10230/58023) for details.
+Check the original [paper](https://arxiv.org/abs/2309.16418) for details.
 
 
 ## Environmental Impact