-
Notifications
You must be signed in to change notification settings - Fork 31
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
1 parent
7cea870
commit e167d27
Showing
1 changed file
with
3 additions
and
1 deletion.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1,2 +1,4 @@ | ||
# SC-LSTM | ||
We focus on text generation, witch is a challenge task that generates a long text under the theme of multiple words. In detail, given a set W = {w1, w2, ..., wk}, this generator aims at generates a text under the semantic information of those words. SC-LSTM ([Wen et al., 2015](http://www.emnlp2015.org/proceedings/EMNLP/pdf/EMNLP199.pdf)) is the best paper of EMNLP 2015, which is is a statistical language generator based on a semantically controlled Long Short-term Memory structure for response generation. The author incorporates a dialogue act 1-hot vector into the original LSTM model and enables the generator to output the word-related text. We directly use this model for our task. And we input a set of words represented by 1-hot vector instead of dialogue act vector in our task. | ||
We focus on text generation, witch is a challenge task that generates a long text under the theme of multiple words. In detail, given a set W = {w1, w2, ..., wk}, this generator aims at generates a text under the semantic information of those words. SC-LSTM ([Wen et al., 2015](http://www.emnlp2015.org/proceedings/EMNLP/pdf/EMNLP199.pdf)) is the best paper of EMNLP 2015, which is is a statistical language generator based on a semantically controlled Long Short-term Memory structure for response generation. The author incorporates a dialogue act 1-hot vector into the original LSTM model and enables the generator to output the word-related text. We directly use this model for our task. And we input a set of words represented by 1-hot vector instead of dialogue act vector in our task. | ||
|
||
The code in this repository is written in Python 2.7/TensorFlow 0.12. And if you use other versions of Python or TensorFlow, you should modify some code. Since SC-LSTM is based on original LSTM, we modify some code based on BasicLSTMCell class of TensorFlow to develop SC-LSTM model (detail in SC_LSTM_Model.py). |