Recent advancements in Large Language Models (LLMs) have significantly improved text generation capabilities, but these systems are still known to hallucinate, and granular uncertainty estimation for long-form LLM generations remains challenging.

Table 6 provides summary statistics of domain coverage.

A recent line of work aims at specially prompting LLMs to review their own generations and generate meaningful natural language feedback, which can subsequently be used to refine them [Madaan et al.,

Code is available at https://github.com/xbyym/DLSR. If the test data does not follow the training distribution, the model could unintentionally produce nonsensical predictions, resulting in some misleading conclusions.

However, computing the second-order Hessian for large-scale models is intractable.