model update

README.md

@@ -46,6 +46,12 @@ model-index:
     - name: MoverScore
       type: moverscore
       value: 0.6587523781250593
+    - name: QAAlignedF1Score (BERTScore)
+      type: qa_aligned_f1_score_bertscore
+      value: 0.9208434218073206
+    - name: QAAlignedF1Score (MoverScore)
+      type: qa_aligned_f1_score_moverscore
+      value: 0.7145302199593226
 ---
 
 # Model Card of `lmqg/mbart-large-cc25-ruquad`
@@ -112,6 +118,13 @@ question = pipe('Нелишним будет отметить, что, разв
 | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | default | 0.188 | 0.342 | 0.293 | 0.872 | 0.659 | [link](https://huggingface.co/lmqg/mbart-large-cc25-ruquad/raw/main/eval/metric.first.sentence.paragraph_answer.question.lmqg_qg_ruquad.default.json) | 
 
 
+### Metrics (QAG)
+
+| Dataset | Type | QA Aligned F1 Score (BERTScore) | QA Aligned F1 Score (MoverScore) | Link |
+|:--------|:-----|--------------------------------:|---------------------------------:|-----:|
+| [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | default | 0.921 | 0.715 | [link](https://huggingface.co/lmqg/mbart-large-cc25-ruquad/raw/main/eval/metric.first.answer.paragraph.questions_answers.lmqg_qg_ruquad.default.json) | 
+    
+
 
 
 ## Training hyperparameters