SPARK: Skeleton-Guided Reasoning Synthesis from Large-Scale Scientific Literature
2026-08-31 • Artificial Intelligence
Artificial Intelligence
AI summaryⓘ
The authors created SPARK to improve how open-source models handle scientific reasoning by focusing on understanding how scientific claims are supported by evidence, rather than just recalling facts or formulas. They designed SPARK to break down research papers into a simplified structure highlighting claims and evidence, and generated reasoning tasks from different scientific angles like mechanisms and hypothesis testing. This approach enabled them to build Spark-234K, a large and challenging dataset that proved more effective than existing ones, even with less training data. Their work aims to help AI better understand and reason about scientific research.
scientific reasoningclaim-evidence structuremechanistic reasoninghypothesis falsificationquantitative derivationboundary calibrationdataset synthesisresearch papersopen-source models
Authors
Yu Li, Wei Li, Xin Gao, Mengyuan Sun, Xiaoyang Wang, Qizhi Pei, Lijun Wu
Abstract
Scientific reasoning remains challenging for open-source models, largely due to the lack of high-quality scientific reasoning data. Existing datasets are often dominated by factual recall or formulaic problem solving, with limited emphasis on mechanism understanding, evidence-grounded reasoning, and hypothesis evaluation. To address this, we introduce SPARK (Scientific Paper Abstracted Reasoning sKeleton), a paper-oriented synthesis framework built on Sci-Base, a large-scale corpus of research papers spanning 10 scientific disciplines. Instead of directly converting papers into question-answer pairs, SPARK treats the claim-evidence-derivation structure of a paper as the fundamental unit of reasoning synthesis. Specifically, SPARK (1) distills each paper into a compact reasoning skeleton capturing its central claims and supporting evidence, enabling self-contained question generation, and (2) synthesizes reasoning tasks from four scientific perspectives: mechanistic reasoning, hypothesis falsification, quantitative derivation, and boundary calibration. A final consistency verification stage further removes unsupported or contradictory outputs. Using this framework, we construct Spark-234K, a scientific reasoning dataset with substantially higher difficulty and diversity than existing resources. Experiments show that Spark-234K consistently outperforms existing scientific reasoning datasets while achieving stronger performance with significantly fewer training samples.