Conformal prediction for text infilling and part-of-speech prediction

TL;DR

This paper introduces conformal prediction algorithms for text infilling and POS tagging, providing statistically reliable confidence sets that are valid in finite samples, demonstrated on the Brown Corpus and real audio transcriptions.

Contribution

It develops new conformal prediction methods integrated with BERT and BiLSTM models for NLP tasks, ensuring valid set predictions with practical applicability.

Findings

Algorithms produce valid confidence sets in simulations

Set sizes are small enough for real-world use

Improved machine transcription accuracy with set predictions

Abstract

Modern machine learning algorithms are capable of providing remarkably accurate point-predictions; however, questions remain about their statistical reliability. Unlike conventional machine learning methods, conformal prediction algorithms return confidence sets (i.e., set-valued predictions) that correspond to a given significance level. Moreover, these confidence sets are valid in the sense that they guarantee finite sample control over type 1 error probabilities, allowing the practitioner to choose an acceptable error rate. In our paper, we propose inductive conformal prediction (ICP) algorithms for the tasks of text infilling and part-of-speech (POS) prediction for natural language data. We construct new conformal prediction-enhanced bidirectional encoder representations from transformers (BERT) and bidirectional long short-term memory (BiLSTM) algorithms for POS tagging and a new conformal prediction-enhanced BERT algorithm for text infilling. We analyze the performance of the algorithms in simulations using the Brown Corpus, which contains over 57,000 sentences. Our results demonstrate that the ICP algorithms are able to produce valid set-valued predictions that are small enough to be applicable in real-world applications. We also provide a real data example for how our proposed set-valued predictions can improve machine generated audio transcriptions.