Toward Ethical AI Through Bayesian Uncertainty in Neural Question Answering
This work addresses the need for more responsible and ethical AI deployment by enhancing interpretability in question-answering systems, though it is incremental as it builds on existing Bayesian techniques.
The paper tackled the problem of uncertainty quantification in neural question answering by applying Bayesian methods to models like LoRA-adapted transformers on the CommonsenseQA benchmark, resulting in improved uncertainty calibration and selective prediction that allows models to abstain when confidence is low.
We explore Bayesian reasoning as a means to quantify uncertainty in neural networks for question answering. Starting with a multilayer perceptron on the Iris dataset, we show how posterior inference conveys confidence in predictions. We then extend this to language models, applying Bayesian inference first to a frozen head and finally to LoRA-adapted transformers, evaluated on the CommonsenseQA benchmark. Rather than aiming for state-of-the-art accuracy, we compare Laplace approximations against maximum a posteriori (MAP) estimates to highlight uncertainty calibration and selective prediction. This allows models to abstain when confidence is low. An ``I don't know'' response not only improves interpretability but also illustrates how Bayesian methods can contribute to more responsible and ethical deployment of neural question-answering systems.