Neural Network of NLI Fail to Capture the General Notion of Inference

• Event: Seminar
• Lecturer: Aarne Talman,
• Date: 08 March 2019
• Duration: 2 hours
• Venue: Gothenburg

Natural language inference (NLI), the task of determining if a sentence is entailed by one or more given sentences, has been a very popular line of research in the NLP community. Due to the popularity and recent advances in neaural network, architectures, significant progress has been made in NLI research, especially with the introduction of various pre-trained contextual language models, like ELMo and BERT. However, there are number of concerns also raised about the current NLI research mostly due to the shortcomings of the current NLI datasets.

In my talk, I will introduce the neural network approaches used in NLI and describe our sentence representation architecture, Hierarchical BiLSTMs (HBMP), which has been successful in many NLI tasks. I will give an overview of some of the criticism and negative results in NLI and show how in our most recent experiments even the pre-trained language models fail to generalize across different NLI datasets.