Ranking Clarification Questions via Natural Language Inference

TL;DR

This paper proposes a novel approach to ranking clarification questions in natural language processing by framing it as a Natural Language Inference task, leveraging BERT models to improve accuracy significantly.

Contribution

It introduces a new NLI-based method for ranking clarification questions and demonstrates substantial performance improvements over existing baselines.

Findings

40-60% improvement in Precision@1 over baselines

Effective use of BERT fine-tuned on NLI datasets

Significant advancement in clarification question ranking accuracy

Abstract

Given a natural language query, teaching machines to ask clarifying questions is of immense utility in practical natural language processing systems. Such interactions could help in filling information gaps for better machine comprehension of the query. For the task of ranking clarification questions, we hypothesize that determining whether a clarification question pertains to a missing entry in a given post (on QA forums such as StackExchange) could be considered as a special case of Natural Language Inference (NLI), where both the post and the most relevant clarification question point to a shared latent piece of information or context. We validate this hypothesis by incorporating representations from a Siamese BERT model fine-tuned on NLI and Multi-NLI datasets into our models and demonstrate that our best performing model obtains a relative performance improvement of 40 percent and 60 percent respectively (on the key metric of Precision@1), over the state-of-the-art baseline(s) on the two evaluation sets of the StackExchange dataset, thereby, significantly surpassing the state-of-the-art.