Towards Faithful and Robust LLM Specialists for Evidence-Based Question-Answering
This work addresses the challenge of making LLM answers more faithful and traceable for research and practical applications, representing an incremental improvement in fine-tuning methods.
The paper tackled the problem of improving source quality and answer attributability in evidence-based question-answering with large language models by introducing a data generation pipeline with automated quality filters and four test sets for benchmarking. The result showed that fine-tuning on synthetic data improved performance on in- and out-of-distribution tasks, with data quality being more critical than quantity.
Advances towards more faithful and traceable answers of Large Language Models (LLMs) are crucial for various research and practical endeavors. One avenue in reaching this goal is basing the answers on reliable sources. However, this Evidence-Based QA has proven to work insufficiently with LLMs in terms of citing the correct sources (source quality) and truthfully representing the information within sources (answer attributability). In this work, we systematically investigate how to robustly fine-tune LLMs for better source quality and answer attributability. Specifically, we introduce a data generation pipeline with automated data quality filters, which can synthesize diversified high-quality training and testing data at scale. We further introduce four test sets to benchmark the robustness of fine-tuned specialist models. Extensive evaluation shows that fine-tuning on synthetic data improves performance on both in- and out-of-distribution. Furthermore, we show that data quality, which can be drastically improved by proposed quality filters, matters more than quantity in improving Evidence-Based QA.