Update README.md

README.md

@@ -90,7 +90,7 @@ It has the same structure as the model introduced as `Prefix LM` in the T5 paper
 
 ### Prefix-LM Model
 
-GPTSAN has the structure of the model named Prefix-LM in the `T5` paper. (The original GPTSAN repository calls it `hybrid`)
+GPTSAN has the structure of the model named Prefix-LM in the [T5 paper](https://arxiv.org/abs/1910.10683). (The original GPTSAN repository calls it `hybrid`)
 In GPTSAN, the `Prefix` part of Prefix-LM, that is, the input position that can be referenced by both tokens, can be specified with any length.
 Arbitrary lengths can also be specified differently for each batch.
 This length applies to the text entered in `prefix_text` for the tokenizer.