SPARC-RAG: Adaptive Sequential-Parallel Scaling with Context Management for Retrieval-Augmented Generation
Published in arXiv preprint, 2026
SPARC-RAG proposes a multi-agent framework that improves retrieval-augmented generation systems’ ability to handle multi-hop questions. The framework coordinates sequential and parallel reasoning through a unified context management mechanism, avoiding the “context pollution” problem. Experiments demonstrate approximately 6.2 F1 score improvement on multiple QA benchmarks while reducing inference costs.
Recommended citation: **Yuxin Yang**, Gangda Deng, Ömer Faruk Akgül, Nima Chitsazan, Yash Govilkar, Akasha Tigalappanavara, Shi-Xiong Zhang, Sambit Sahu, Viktor Prasanna. "SPARC-RAG: Adaptive Sequential-Parallel Scaling with Context Management for Retrieval-Augmented Generation." arXiv preprint arXiv:2602.00083, 2026.
Download Paper