{"type":"ScholarlyArticle","title":"ScholarSum: Student-Teacher Abstractive Summarization via Knowledge Graph Reasoning and Reflective Refinement","authors":[{"name":"Bohou Zhang","slug":"bohou-zhang","given":"Bohou","family":"Zhang","affiliation":""},{"name":"Xiaoyu Tao","slug":"xiaoyu-tao","given":"Xiaoyu","family":"Tao","affiliation":""},{"name":"Mingyue Cheng","slug":"mingyue-cheng","given":"Mingyue","family":"Cheng","affiliation":""},{"name":"Huijie Liu","slug":"huijie-liu","given":"Huijie","family":"Liu","affiliation":""},{"name":"Qi Liu","slug":"qi-liu","given":"Qi","family":"Liu","affiliation":""}],"abstract":"Abstractive summarization plays a crucial role in enabling efficient understanding of scientific literature, yet it inherently demands both linguistic fluency and factual faithfulness. Existing approaches often fail to reconcile these two requirements. Extractive methods rely on rigid sentence splicing that disrupts macro-level logical coherence, while large language model (LLM)-based generative approaches, despite mastering linguistic fluency, exhibit limited factual consistency.\r\nIn this work, we propose ScholarSum, a hierarchical reflective graph-based framework that emulates a student–teacher writing process for fluent and faithful scientific summarization. ScholarSum first organizes the document into a hierarchical knowledge graph by segmenting it into semantically coherent units, whose multi-layered community structure captures global logic and macro-level themes. Guided by this global structure, the student generates an initial draft, which is subsequently refined through fine-grained evidence retrieval.\r\nTo ensure factual consistency, a teacher-like reviewer then iteratively examines the draft, identifies unsupported content, and prompts targeted re-retrieval and rewriting until the summary meets rigorous quality standards.\r\nExtensive experiments demonstrate that ScholarSum significantly outperforms previous baselines in terms of both completeness and faithfulness. \r\nOur code is available at https://github.com/Xiaoyu-Tao/ScholarSum.","doi":"10.24963/ijcai.2026/675","url":"https://ijcai.org/proceedings/2026/675","pdf":"https://ijcai.org/proceedings/2026/0675.pdf","markdown":"https://ijcai.org/proceedings/2026/675.md","bibtex":"https://ijcai.org/proceedings/2026/bibtex/675","volume":{"slug":"2026","title":"Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2026","url":"https://ijcai.org/proceedings/2026/","year":2026,"location":"Bremen, Germany","dates":"15-21 August 2026","editors":"Diego Calvanese","isbn":"978-1-956792-09-6","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2026"},"section":"Natural Language Processing","track":"Main Track","number":675,"pages":{"first":6064,"last":6072},"keywords":["Natural Language Processing: Information retrieval and text mining"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}