Week One: 9/4 – 9/8

Natural language sentence matching (NLSM)
is the task of comparing two sentences and identifying the relationship between them.”


BiMPM: Bilateral Multi-Perspective Matching for Natural Language Sentences

For the first week of our research project, my task was to read several papers relating to hedge cues and BiMPM, a tool that we will use to successfully execute our project. The papers I’ve read includes Maria Georgescul’s A hedgehop over a max-margin framework using hedge cues which describes the experimental settings adopted for detecting sentences containing uncertainty, and also Zhiguo Wang, Wael Hamza, and Radu Florian’s Bilateral Multi-Perspective Matching for Natural Language Sentences which talks about using the BiMPM model for NLSM tasks.

How it works:
Given two sentences P and Q, our model first encodes them with a bidirectional Long ShortTerm Memory Network (BiLSTM) encoder. Next, we match the two encoded sentences in two directions P against Q and Q against P. In each matching direction, each time step of one sentence is matched against all timesteps of the other sentence from multiple perspectives. Then, another BiLSTM layer is utilized to aggregate the matching results into a fixed-length matching vector. Finally, based on the matching vector, a decision is made through a fully connected layer.

I was also tasked to clone this repository and familiarize myself on how to run it on the server that we will be using.