ScholarStack：面向科学智能体的分层研究资产编排与跨任务复用
ScholarStack: Layered Research Asset Orchestration and Cross-Task Reuse for Scientific Agents
ScholarSeed AI 团队提出 ScholarStack，一个将论文集合编译为可复用、带版本与溯源信息资产的分层研究资产框架，覆盖论文级陈述、领域级组织和跨论文证据综合三个层次。该框架在四个任务族共十个任务设置上实例化，使用编译资产的智能体在跨论文证据任务（如多论文问答与文献综述生成）上取得质量提升，且每个被测量的任务查询期 token 成本均下降，资产一次编译即可跨任务复用。
Authors:ScholarSeed AI Team: Caoqinwei Gong, Xue Jiang, Wei Luo, Xiaoyu Qiu, Jiayi Sheng, Yi Wang, Zheng Yu, Ao Zhang, Haifan Zhang, Hanwei Zhang, Jihai Zhang, Yuan Cao, Wei Chen, Liyun Dai, Wenkai Fang, Guanglei Wang, Kai Ying, Tingyu Zhu, Wotao Yin
Abstract:Scientific agents support a range of literature-based research tasks, such as retrieval, question answering, evidence-grounded generation, and claim assessment. Most existing systems, however, are organized around individual tasks: the same papers are repeatedly retrieved, segmented, and interpreted, and the understanding built in one task is difficult to reuse in the next. We present ScholarStack, a layered research asset framework that compiles a paper collection into reusable, versioned, and provenance-preserving assets at three complementary levels: source-grounded paper-level statements, domain-level organization, and evidence-grounded cross-paper syntheses. A common access interface returns task-specific views at the evidence granularity each task requires, preserving study conditions, source traceability, and verification status. We instantiate the framework on four task families spanning ten task settings, comparing agents that use the compiled assets with task-specific baselines under matched base models. Quality gains concentrate on tasks that require cross-paper evidence, such as multi-paper question answering and literature review generation, and query-time token cost falls on every task where it is measured, with assets compiled once and reused across tasks. These results suggest that layered research assets can serve as shared infrastructure for scientific agents, shifting literature-based assistance from isolated document processing toward cumulative, evidence-grounded workflows.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2609.23735 [cs.AI] |
| (or arXiv:2609.23735v3 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2609.23735 arXiv-issued DOI via DataCite |
Submission history
From: Wei Luo [view email]
[v1]
Sun, 20 Sep 2026 16:25:06 UTC (1,094 KB)
[v2]
Tue, 22 Sep 2026 06:18:25 UTC (1,094 KB)
[v3]
Mon, 12 Oct 2026 13:49:45 UTC (1,271 KB)
来源：arXiv：cs.AI · arxiv.org