Update README.md

README.md

@@ -9,7 +9,7 @@ library_name: transformers
 license: mit
 ---
 
-## Model description: deberta-v3-large-zeroshot-v1.1-all-33
+# Model description:  deberta-v3-large-zeroshot-v1.1-all-33
 The model is designed for zero-shot classification with the Hugging Face pipeline. 
 
 The model can do one universal task: determine whether a hypothesis is `true` or `not_true`
@@ -17,10 +17,12 @@ given a text (also called `entailment` vs. `not_entailment`).
 This task format is based on the Natural Language Inference task (NLI).
 The task is so universal that any classification task can be reformulated into the task.
 
+A detailed description of how the model was trained and how it can be used is available in this paper: [link to be added]
+
 ## Training data
-The model was trained on a mixture of 33 datasets and 389 classes that have been reformatted into this universal format.
+The model was trained on a mixture of __33 datasets and 387 classes__ that have been reformatted into this universal format.
 1.   Five NLI datasets with ~885k texts: "mnli", "anli", "fever", "wanli", "ling"
-2.   28 classification tasks with ~51k texts:
+2.   28 classification tasks reformatted into the universal NLI format. ~51k cleaned texts were used to avoid overfitting:
 'amazonpolarity', 'imdb', 'appreviews', 'yelpreviews', 'rottentomatoes',
 'emotiondair', 'emocontext', 'empathetic',
 'financialphrasebank', 'banking77', 'massive',
@@ -35,6 +37,10 @@ See details on each dataset here: https://github.com/MoritzLaurer/zeroshot-class
 Note that compared to other NLI models, this model predicts two classes (`entailment` vs. `not_entailment`)
 as opposed to three classes (entailment/neutral/contradiction)
 
+The model was only trained on English data. For __multilingual use-cases__, 
+I recommend machine translating texts to English with libraries like [EasyNMT](https://github.com/UKPLab/EasyNMT).
+English-only models tend to perform better than multilingual models and
+validation with English data can be easier if you don't speak all languages in your corpus.
 
 ### How to use the model
 #### Simple zero-shot classification pipeline