Update README.md

README.md

@@ -6,7 +6,7 @@ datasets:
 metrics:
 - rouge
 model-index:
-- name: eval-mbart-large
+- name: mbart-large-turkish-sum
   results:
   - task:
       name: Summarization
@@ -21,33 +21,20 @@ model-index:
       value: 46.7011
 ---
 
-<!-- This model card has been generated automatically according to the information the Trainer had access to. You
-should probably proofread and complete it, then remove this comment. -->
+# [Mukayese: Turkish NLP Strikes Back](https://arxiv.org/abs/2203.01215)
 
-# eval-mbart-large
+## Summarization: mukayese/mbart-large-turkish-sum
 
 This model is a fine-tuned version of [facebook/mbart-large-50](https://huggingface.co/facebook/mbart-large-50) on the mlsum tu dataset.
+
 It achieves the following results on the evaluation set:
-- Loss: 2.8386
+
 - Rouge1: 46.7011
 - Rouge2: 34.0087
 - Rougel: 41.5475
 - Rougelsum: 43.2108
-- Gen Len: 43.2426
-
-## Model description
-
-More information needed
-
-## Intended uses & limitations
-
-More information needed
 
-## Training and evaluation data
-
-More information needed
-
-## Training procedure
+Check [this](https://arxiv.org/abs/2203.01215) paper for more details on the model and the dataset.
 
 ### Training hyperparameters
 
@@ -67,25 +54,22 @@ The following hyperparameters were used during training:
 - mixed_precision_training: Native AMP
 - label_smoothing_factor: 0.1
 
-### Training results
-
-| Training Loss | Epoch | Step  | Validation Loss | Rouge1  | Rouge2  | Rougel  | Rougelsum | Gen Len |
-|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
-| 2.9098        | 1.0   | 3895  | 2.8820          | 45.2085 | 33.0253 | 40.3511 | 41.8378   | 40.1802 |
-| 2.7496        | 2.0   | 7790  | 2.8620          | 45.3455 | 33.0049 | 40.4574 | 41.9738   | 40.845  |
-| 2.6215        | 3.0   | 11685 | 2.8386          | 46.6642 | 34.0133 | 41.5102 | 43.1852   | 43.3505 |
-| 2.5031        | 4.0   | 15580 | 2.8620          | 46.5081 | 33.9028 | 41.5001 | 43.0841   | 42.4534 |
-| 2.3967        | 5.0   | 19475 | 2.8935          | 46.1029 | 33.4495 | 41.0557 | 42.7096   | 41.9169 |
-| 2.3161        | 6.0   | 23370 | 2.9255          | 46.0193 | 33.2904 | 40.9323 | 42.575    | 42.5379 |
-| 2.2348        | 7.0   | 27265 | 2.9692          | 46.4242 | 33.718  | 41.4037 | 43.0504   | 41.7957 |
-| 2.1716        | 8.0   | 31160 | 3.0044          | 46.1669 | 33.3276 | 40.9307 | 42.7015   | 42.8942 |
-| 2.1179        | 9.0   | 35055 | 3.0372          | 46.0666 | 33.2483 | 40.9372 | 42.6837   | 42.7636 |
-| 2.0753        | 10.0  | 38950 | 3.0627          | 46.1444 | 33.2551 | 40.9514 | 42.7096   | 42.9266 |
-
-
 ### Framework versions
 
 - Transformers 4.11.3
 - Pytorch 1.8.2+cu111
 - Datasets 1.14.0
 - Tokenizers 0.10.3
+
+### Citation
+
+```
+@misc{safaya-etal-2022-mukayese,
+    title={Mukayese: Turkish NLP Strikes Back},
+    author={Ali Safaya and Emirhan Kurtuluş and Arda Göktoğan and Deniz Yuret},
+    year={2022},
+    eprint={2203.01215},
+    archivePrefix={arXiv},
+    primaryClass={cs.CL}
+}
+```