On Context Utilization in Summarization with Large Language Models
This addresses a critical limitation in summarization for users relying on LLMs to process long documents, though it is incremental as it builds on known biases from question answering.
The paper tackles the problem of position bias in large language models (LLMs) for summarization, where models unevenly utilize input context, favoring initial and final segments, and introduces a new benchmark called MiddleSum to evaluate this issue, showing that alternative inference methods like hierarchical and incremental summarization can mitigate the bias.
Large language models (LLMs) excel in abstractive summarization tasks, delivering fluent and pertinent summaries. Recent advancements have extended their capabilities to handle long-input contexts, exceeding 100k tokens. However, in question answering, language models exhibit uneven utilization of their input context. They tend to favor the initial and final segments, resulting in a U-shaped performance pattern concerning where the answer is located within the input. This bias raises concerns, particularly in summarization where crucial content may be dispersed throughout the source document(s). Besides, in summarization, mapping facts from the source to the summary is not trivial as salient content is usually re-phrased. In this paper, we conduct the first comprehensive study on context utilization and position bias in summarization. Our analysis encompasses 6 LLMs, 10 datasets, and 5 evaluation metrics. We introduce a new evaluation benchmark called MiddleSum on the which we benchmark two alternative inference methods to alleviate position bias: hierarchical summarization and incremental summarization. Our code and data can be found here: https://github.com/ntunlp/MiddleSum.