By reading source code, I think this project reduce noise impact in distant supervision data set is by use sentence-level attention, rather than multi-instance learning.
see code as follow:
This section is forward() in selector.py, it dose not chose the most reliable sentence as what is mentioned in multi-instance learning paper. so am I right?
By reading source code, I think this project reduce noise impact in distant supervision data set is by use sentence-level attention, rather than multi-instance learning. see code as follow:
This section is
forward()
inselector.py
, it dose not chose the most reliable sentence as what is mentioned in multi-instance learning paper. so am I right?Appreciate for your help.