• Publications
  • Influence
ParsiNLU: A Suite of Language Understanding Challenges for Persian
TLDR
This work introduces ParsiNLU, the first benchmark in Persian language that includes a range of language understanding tasks—reading comprehension, textual entailment, and so on, and presents the first results on state-of-the-art monolingual and multilingual pre-trained language models on this benchmark and compares them with human performance.
Rating for Parents: Predicting Children Suitability Rating for Movies Based on Language of the Movies
TLDR
This paper proposes an RNN based architecture with attention that jointly models the genre and the emotions in the script to predict the MPAA rating and achieves 78% weighted F1-score for the classification model that outperforms the traditional machine learning method by 6%.
Community Structure and Information Cascade in Signed Networks
TLDR
This paper investigates influence of community structure (i.e., percentage of inter-community positive and intra-community negative links) on the cascade depth and finds significant influence ofcommunity structure on cascade depth in both model and real networks.
Attending the Emotions to Detect Online Abusive Language
TLDR
A new corpus for the task of abusive language detection is presented that is collected from a semi-anonymous online platform, and unlike the majority of other available resources, is not created based on a specific list of bad words.
Age Suitability Rating: Predicting the MPAA Rating Based on Movie Dialogues
TLDR
This paper creates a corpus for movie MPAA ratings and proposes an RNN based architecture with attention that jointly models the genre and the emotions in the script to predict the MPAA rating.
White Paper-Creating a Repository of Objectionable Online Content: Addressing Undesirable Biases and Ethical Considerations
Executive Summary This white paper summarizes the authors' structured brainstorming regarding ethical considerations for creating an extensive repository of online content labeled with tags that
From None to Severe: Predicting Severity in Movie Scripts
TLDR
The experimental results show that the method outperforms the previous state-of-the-art model and provides use-ful information to interpret model predictions.
White Paper: Challenges and Considerations for the Creation of a Large Labelled Repository of Online Videos with Questionable Content
TLDR
This white paper presents a summary of the discussions regarding critical considerations to develop an extensive repository of online videos annotated with labels indicating questionable content and what actions to reduce risk of trauma to annotators.
A Case Study of Deep Learning-Based Multi-Modal Methods for Labeling the Presence of Questionable Content in Movie Trailers
TLDR
This work introduces a new dataset containing videos of movie trailers in English downloaded from IMDB and YouTube, along with their corresponding age-suitability rating labels, and proposes a multi-modal deep learning pipeline addressing the movie trailer age suitability rating problem.
...
...