Year: 2026
Type: conference-paper
Abstract: Recent advances in summarization research focus on improving summary quality across multiple criteria, such as completeness, conciseness, and faithfulness, by jointly optimizing these dimensions.However, these efforts largely overlook the challenge of controlling summary generation with respect to individual criteria, especially in the presence of their inherent trade-offs.For example, enhancing conciseness can compromise completeness, and vice versa.In this work, we address this gap by proposing a loss function that aligns model outputs with fine-grained, model-based evaluation scores (e.g., from FineSurE), enabling both improvement in summary quality and dimensionspecific control.Our approach 1 improves the overall quality of the summaries while maintaining the ability to selectively prioritize one criterion over others.Experiments on three pretrained models (LLaMA, Qwen, and Mistral) demonstrate that our method achieves performance comparable to state-of-the-art summarizers, while uniquely offering strong controllability over individual quality dimensions.
Authors: Hongye Liu, Liang Ding, Ricardo Henao
Institutions: Duke University, The University of Sydney
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7166842593