Introduction: With social media continuing to grow as the cultural juggernaut it has become in the past two decades, the amount of information the average person has at their fingertips is much higher than it previously has been. As such, it becomes essential for individuals to have a quick way to verify the information they read online and to have a check to ensure they are not spreading false information. This is especially necessary in the field of public health, where misinformation about COVID-19 could put individuals at risk for consequences to their health. The Fax&Knowledge FactChecker will act as a convenient tool to assess the validity of claims online, querying the relevant contradictory information from a reputable source. The paper that we will be implementing focuses on creating an effective end-to-end fact checker using a solely a language model, with-out any external knowledge or explicit retrieval components. We chose this paper because it shows how with a good masking mechanism and verification model, language models can be used for fact checking. Challenges: The hardest part of the project is integrating our individual parts of our model together, such as the preprocessing of our data and the Named Entity Recognition model . Additionally, implementing the BERT model has been particularly challenging. Initializing the model ran a lot of errors. Setting up the model and matching the inputs to our preprocessing outputs was a difficult task due to the limitations of our data set. Insights: No concrete results up to this point, since we have implemented different parts of the model independently and we have yet to connect them with each other. Plan: We are on track to finish our project. We need to dedicate more time on setting up our evaluation metric. Currently, we are considering whether we should implement more naive parts of our end-to-end model in order to compare against the pre-trained models.
Log in or sign up for Devpost to join the conversation.