Uncertainty Estimation in Autoregressive Structured Prediction

TL;DR

This paper introduces a unified probabilistic ensemble framework for estimating uncertainty in autoregressive structured prediction tasks, addressing both theoretical and practical challenges across sequence data.

Contribution

It proposes a novel ensemble-based approach for uncertainty estimation in structured prediction, filling a gap in existing research focused mainly on unstructured tasks.

Findings

Provides baselines for error detection at token and sequence levels.

Demonstrates effectiveness on translation and speech recognition datasets.

Discusses interpretability and application of uncertainty measures.

Abstract

Uncertainty estimation is important for ensuring safety and robustness of AI systems. While most research in the area has focused on un-structured prediction tasks, limited work has investigated general uncertainty estimation approaches for structured prediction. Thus, this work aims to investigate uncertainty estimation for autoregressive structured prediction tasks within a single unified and interpretable probabilistic ensemble-based framework. We consider: uncertainty estimation for sequence data at the token-level and complete sequence-level; interpretations for, and applications of, various measures of uncertainty; and discuss both the theoretical and practical challenges associated with obtaining them. This work also provides baselines for token-level and sequence-level error detection, and sequence-level out-of-domain input detection on the WMT'14 English-French and WMT'17 English-German translation and LibriSpeech speech recognition datasets.