Calibrating Large Language Models Using Their Generations Only
This addresses the challenge of building trust and safety in LLM deployments by enabling confidence estimation without model access, though it is incremental as it builds on existing calibration techniques.
The authors tackled the problem of calibrating large language models (LLMs) when only their generated text is available, proposing APRICOT, a method that trains an auxiliary model to predict confidence from text, and demonstrated competitive calibration error on closed-book question-answering tasks.
As large language models (LLMs) are increasingly deployed in user-facing applications, building trust and maintaining safety by accurately quantifying a model's confidence in its prediction becomes even more important. However, finding effective ways to calibrate LLMs - especially when the only interface to the models is their generated text - remains a challenge. We propose APRICOT (auxiliary prediction of confidence targets): A method to set confidence targets and train an additional model that predicts an LLM's confidence based on its textual input and output alone. This approach has several advantages: It is conceptually simple, does not require access to the target model beyond its output, does not interfere with the language generation, and has a multitude of potential usages, for instance by verbalizing the predicted confidence or adjusting the given answer based on the confidence. We show how our approach performs competitively in terms of calibration error for white-box and black-box LLMs on closed-book question-answering to detect incorrect LLM answers.