Add evaluation results on the default config of xsum

README.md

@@ -41,6 +41,39 @@ model-index:
       type: gen_len
       value: 78.5866
       verified: true
+  - task:
+      type: summarization
+      name: Summarization
+    dataset:
+      name: xsum
+      type: xsum
+      config: default
+      split: test
+    metrics:
+    - name: ROUGE-1
+      type: rouge
+      value: 20.791
+      verified: true
+    - name: ROUGE-2
+      type: rouge
+      value: 3.6084
+      verified: true
+    - name: ROUGE-L
+      type: rouge
+      value: 13.607
+      verified: true
+    - name: ROUGE-LSUM
+      type: rouge
+      value: 16.6433
+      verified: true
+    - name: loss
+      type: loss
+      value: 2.5788395404815674
+      verified: true
+    - name: gen_len
+      type: gen_len
+      value: 67.6076
+      verified: true
 ---
 # BART (large-sized model), fine-tuned on CNN Daily Mail