Scholarstack enables reuse of scientific paper analyses across tasks
ScholarStack: Layered Research Asset Orchestration and Cross-Task Reuse for Scientific Agents
Artificial Intelligence
Summary
Scientific work often requires reading many papers to answer questions or summarize findings, but existing tools treat each task separately, repeatedly processing the same papers. The authors introduce ScholarStack, a system that organizes information from papers into layers that can be reused for different tasks, saving time and effort. Their tests show ScholarStack helps tasks needing evidence from multiple papers perform better and reduces the computational cost of repeated analysis. This approach moves scientific AI tools toward more integrated and efficient research workflows.
What this means in practice
- •For scientific software developers: Implement research assistants that efficiently reuse paper analyses across tasks to reduce redundant processing and improve evidence integration.
- •For legal document analysts: Create tools that compile legal documents into layered reusable assets to support multiple research and question answering tasks efficiently.
Authors
ScholarSeed AI Team, Ao Zhang, Caoqinwei Gong, Guanglei Wang, Haifan Zhang, Hanwei Zhang, Jiayi Sheng, Jihai Zhang, Kai Ying, Liyun Dai, Tingyu Zhu, Wei Chen, Wei Luo, Wenkai Fang, Xiaoyu Qiu, Xue Jiang, Yi Wang, Yuan Cao, Zheng Yu, 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.