diff --git a/docs/source/model_summary.mdx b/docs/source/model_summary.mdx
index 4542f541e1..215b572f60 100644
--- a/docs/source/model_summary.mdx
+++ b/docs/source/model_summary.mdx
@@ -618,6 +618,9 @@ The library provides a version of this model for conditional generation.
+
+
+
[Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://arxiv.org/abs/1910.10683), Colin Raffel et al.