"None of the Above":Measure Uncertainty in Dialog Response Retrieval
This addresses uncertainty estimation for dialog systems, but it is incremental as it builds on existing retrieval models without introducing new methods.
The paper tackled the problem of measuring uncertainty in dialog response retrieval by showing that a model's confidence can be captured without retraining, achieving results on the Ubuntu Dialog Corpus with minimal computational overhead.
This paper discusses the importance of uncovering uncertainty in end-to-end dialog tasks, and presents our experimental results on uncertainty classification on the Ubuntu Dialog Corpus. We show that, instead of retraining models for this specific purpose, the original retrieval model's underlying confidence concerning the best prediction can be captured with trivial additional computation.