IDEAL: Leveraging Infinite and Dynamic Characterizations of Large Language Models for Query-focused Summarization

TL;DR

This paper introduces IDEAL, a novel approach for query-focused summarization using large language models, emphasizing infinite and dynamic characterizations to improve relevance and efficiency.

Contribution

The paper proposes two new modules, Query-aware HyperExpert and Query-focused Infini-attention, to enhance LLM-based QFS by leveraging key characteristics for better summarization.

Findings

Effective on multiple QFS benchmarks

Demonstrates improved relevance and efficiency

Generalizes well across different datasets

Abstract

Query-focused summarization (QFS) aims to produce summaries that answer particular questions of interest, enabling greater user control and personalization. With the advent of large language models (LLMs), shows their impressive capability of textual understanding through large-scale pretraining, which implies the great potential of extractive snippet generation. In this paper, we systematically investigated two indispensable characteristics that the LLMs-based QFS models should be harnessed, Lengthy Document Summarization and Efficiently Fine-grained Query-LLM Alignment, respectively. Correspondingly, we propose two modules called Query-aware HyperExpert and Query-focused Infini-attention to access the aforementioned characteristics. These innovations pave the way for broader application and accessibility in the field of QFS technology. Extensive experiments conducted on existing QFS benchmarks indicate the effectiveness and generalizability of the proposed approach. Our code is publicly available at https://github.com/DCDmllm/IDEAL_Summary.