Non-Exchangeable Conformal Language Generation with Nearest Neighbors

Anonymous

Abstract: Quantifying uncertainty in automatically generated text is important for letting humans check potential hallucinations and making systems more reliable. Conformal prediction is an attractive framework to provide predictions imbued with statistical guarantees, however, its application to text generation is challenging since any i.i.d. assumptions are not realistic. In this paper, we bridge this gap by leveraging recent results on non-exchangeable conformal prediction, which still ensures bounds on coverage. The result is a novel extension of the conformal prediction framework to generation based on nearest neighbors. Our method can be used post-hoc for an arbitrary model without extra training and supplies token-level, calibrated prediction sets equipped with statistical guarantees. Experiments in machine translation and language modeling show encouraging results in word coverage and generation quality.

Paper Type: long

Research Area: Machine Learning for NLP

Contribution Types: NLP engineering experiment

Languages Studied: English, German, Japanese

Consent To Share Submission Details: On behalf of all authors, we agree to the terms above to share our submission details.