发表机构
School of Future Technology, South China University of Technology; TikTok Inc; School of Artificial Intelligence, Sun Yat-sen University; School of Software Engineering, Sun Yat-sen University（华南理工大学未来技术学院; 字节跳动公司; 中山大学人工智能学院; 中山大学软件工程学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对现有RAG方法的认知孤岛化与跨层证据脱节问题，提出受脑桥启发的PonsRAG框架，通过三层索引与协调推理组件，在四项长文本叙事基准上实现多项选择任务平均准确率11.56%的相对提升。
AI 中文摘要
长文本推理是处理和推理复杂叙事的核心能力。尽管检索增强生成（RAG）提供了有前景的框架，但现有方法仍面临两个关键挑战：认知孤岛化和跨层证据脱节。为解决这些问题，我们提出受生物脑桥（pons）启发的协调RAG框架PonsRAG。该框架包含两个关键组件：一是三层索引，将文档组织成连通的知识结构以连接认知孤岛；二是协调推理，可在不同层间检索证据并将跨层信息整合为统一上下文。我们在四个长上下文叙事基准上对PonsRAG进行评估，实验结果显示其优于最强基线，在多项选择任务的平均准确率上实现了11.56%的相对提升。
英文摘要
Long Narrative Reasoning is an essential capability for processing and reasoning over complex narratives. While retrieval-augmented generation provides a promising framework, existing methods still face two critical challenges: cognitive islanding and cross-layer evidence disconnection. To address these issues, we propose PonsRAG, a coordinated RAG framework inspired by the biological pons. PonsRAG consists of two key components: Triple-Layer Indexing, which organizes documents into a connected knowledge structure to bridge cognitive islands, and Coordinated Reasoning, which retrieves evidence across distinct layers and integrates cross-layer information into a unified context. We evaluate PonsRAG on four long-context narrative benchmarks, and experimental results show that it outperforms the strongest baseline, achieving a 11.56% relative improvement in average accuracy on multi-choice tasks.
CommentsAccepted to EMNLP 2026 (Main Conference)
