Create README.md
#1
by
duke2502
- opened
README.md
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
We present a novel training paradigm for neural abstractive summarization. Instead of using MLE training alone, we introduce a contrastive learning component, which encourages the abstractive models to estimate the probability of system-generated summaries more accurately.
|