The Year in AI Papers
The year/Independent research

Paper 2512.17220

Mindscape-Aware Retrieval Augmented Generation for Improved Long Context Understanding

Yuqing Li, Jiangnan Li, Zheng Lin, Ziyan Zhou, Junjie Wu, Weiping Wang, Jie Zhou, Mo Yu

Published
Dec 2025
Research lab
Independent
Citations
6
GitHub
Not linked

01 In brief

Summary

The paper introduces Mindscape-Aware RAG (MiA-RAG), a framework that equips LLM-based RAG systems with a global semantic representation, called a mindscape, to improve long-context understanding.

The mindscape is built via hierarchical summarization of a document.

MiA-RAG conditions both retrieval and generation on this mindscape: a retriever (MiA-Emb) integrates the summary into query embeddings, and a generator (MiA-Gen) uses the summary to reason over retrieved chunks.

Evaluated on five long-context benchmarks (NarrativeQA, ∞Bench, DetectiveQA-Zh/En, NoCha), MiA-RAG-14B outperforms vanilla RAG-14B by +16.18% and vanilla RAG-72B by +8.63% in average scores.

Ablations show that removing the summary degrades performance, and analyses reveal that the mindscape aligns query embeddings with the document's semantic space and guides attention to relevant evidence.

The framework also scales across model sizes and is robust to summary quality.

02 From the paper

Abstract

Humans understand long and complex texts by relying on a holistic semantic representation of the content. This global view helps organize prior knowledge, interpret new information, and integrate evidence dispersed across a document, as revealed by the Mindscape-Aware Capability of humans in psychology. Current Retrieval-Augmented Generation (RAG) systems lack such guidance and therefore struggle with long-context tasks. In this paper, we propose Mindscape-Aware RAG (MiA-RAG), the first framework to formulate mindscape-aware retrieval and generation as a unified conditioning paradigm for LLM-based RAG. MiA-RAG builds a mindscape through hierarchical summarization and conditions both retrieval and generation on this global semantic representation. This enables the retriever to form enriched query embeddings and the generator to reason over retrieved evidence within a coherent global context. We evaluate MiA-RAG across diverse long-context and bilingual benchmarks for evidence-based understanding and global sense-making. It consistently surpasses baselines, and further analysis shows that it aligns local details with a coherent global representation, enabling more human-like long-context retrieval and reasoning.