Work

SCURank: Ranking Multiple Candidate Summaries with Summary Content Units for Enhanced Summarization

Year: 2026

Type: conference-paper

Abstract: Small language models (SLMs), such as BART, can achieve summarization performance comparable to large language models (LLMs) via distillation.However, existing LLM-based ranking strategies for summary candidates suffer from instability, while classical metrics (e.g., ROUGE) are insufficient to rank highquality summaries.To address these issues, we introduce SCURank, a framework that enhances summarization by leveraging Summary Content Units (SCUs).Instead of relying on unstable comparisons or surface-level overlap, SCURank evaluates summaries based on the richness and semantic importance of information content.We investigate the effectiveness of SCURank in distilling summaries from multiple diverse LLMs.Experimental results demonstrate that SCURank outperforms traditional metrics and LLM-based ranking methods across evaluation measures and datasets.Furthermore, our findings show that incorporating diverse LLM summaries enhances model abstractiveness and overall distilled model performance, validating the benefits of informationcentric ranking in multi-LLM distillation.The code for SCURank is available at https:// github.com/IKMLab/SCURank.

Source: Findings of the Association for Computational Linguistics: ACL (Findings of ACL)

DOI: 10.18653/v1/2026.findings-acl.1941

Authors: Bo-Jyun Wang, Ying-Jia Lin, Hung-Yu Kao

Institutions: Chang Gung University, National Cheng Kung University, National Tsing Hua University

Language: English

Topic: Topic Modeling

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funder: National Science and Technology Council

Award: G5431289886