Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
To resolve the token inefficiency and fragmented retrieval of agentic search over flat document collections, the authors introduce CorpusMap, an offline navigation layer that organizes corpora around recurring entities. By resolving cross-document entity mentions ahead of time into structured "Entity Pages" linked to source files, the system exposes an entity-document graph for agents to traverse during multi-hop reasoning. Across seven LLMs and three benchmarks, CorpusMap significantly improved evidence discovery and answer accuracy over flat-corpus baselines and four alternative navigation structures while reducing average token consumption.
LLM agents waste immense token budgets repeatedly rediscovering identical cross-document links on every query—pre-indexing corpora into an offline entity-document graph solves fragmented multi-hop retrieval while actually lowering token costs.
Answering questions and completing tasks over large document collections often requires connecting evidence spread across multiple documents, such as a project's approval recorded in one, its requirements in another, and its latest status in a third. Recent LLM agents approach this by iteratively searching the full corpus rather than reading only a fixed set of top-ranked documents. However, when the corpus is exposed only as a flat collection of files, a relevant document gives no indication of how it relates to others, so the agent must rediscover these relationships for every query, often missing complementary evidence while simultaneously consuming substantial additional tokens. To address this, we introduce CorpusMap, a navigation layer that organizes the corpus around its recurring entities, which are identifiable from the documents themselves and can link a single document to many others across sources. Specifically, CorpusMap represents each recurring entity as an Entity Page that aggregates information about it and links to every document that refers to it, forming a graph between entities and documents that the agent can traverse to gather otherwise disconnected evidence. Moreover, since CorpusMap is constructed offline by resolving mentions of the same entity across documents, its links are shared across queries rather than rediscovered repeatedly at inference time. Using 7 different models with 3 benchmark datasets, we show that CorpusMap improves both evidence discovery and answer quality over raw-corpus agentic search while using fewer tokens on average, and further outperforms 4 alternative navigation layers, suggesting that entities serve as effective anchors for navigating large document collections.