Ranking Clarification Questions via Natural Language Inference
This work addresses the need for better machine comprehension in interactive systems like QA forums, though it is incremental as it builds on existing NLI methods.
The paper tackles the problem of ranking clarification questions for natural language queries by framing it as a Natural Language Inference task, achieving relative improvements of 40% and 60% in Precision@1 over state-of-the-art baselines on StackExchange datasets.
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.