diff --git a/examples/benchmarks/ALSTM/README.md b/examples/benchmarks/ALSTM/README.md
index cd9dd3493..1b749bd80 100644
--- a/examples/benchmarks/ALSTM/README.md
+++ b/examples/benchmarks/ALSTM/README.md
@@ -2,9 +2,7 @@
 
 - ALSTM contains a temporal attentive aggregation layer based on normal LSTM.
 
-- The code used in Qlib is a pyTorch implementation of Code: https://github.com/fulifeng/Adv-ALSTM
-
 - Paper: A dual-stage attention-based recurrent neural network for time series prediction.
 
-  https://www.ijcai.org/Proceedings/2017/0366.pdf
+  [https://www.ijcai.org/Proceedings/2017/0366.pdf](https://www.ijcai.org/Proceedings/2017/0366.pdf)
 
diff --git a/examples/benchmarks/SFM/README.md b/examples/benchmarks/SFM/README.md
index eb1c8b157..5f74c15d2 100644
--- a/examples/benchmarks/SFM/README.md
+++ b/examples/benchmarks/SFM/README.md
@@ -1,4 +1,3 @@
 # State-Frequency-Memory
 - State Frequency Memory (SFM) is a novel recurrent network that uses Discrete Fourier Transform to decompose the hidden states of memory cells and capture the multi-frequency trading patterns from past market data to make stock price predictions. 
-- The code used in Qlib is a pyTorch implementation of SFM.
 - Paper: Stock Price Prediction via Discovering Multi-Frequency Trading Patterns. [https://www.cs.ucf.edu/~gqi/publications/kdd2017_stock.pdf.](https://www.cs.ucf.edu/~gqi/publications/kdd2017_stock.pdf.)
\ No newline at end of file