Context Cartography: Toward Structured Governance of Contextual Space in Large Language Model Systems

TL;DR

This paper introduces Context Cartography, a formal framework for managing the structure of contextual information in large language models, addressing limitations of expanding context windows and analyzing current systems.

Contribution

It formalizes a tripartite zonal model and seven cartographic operators for structured governance of contextual space in LLMs, grounded in transformer attention salience geometry.

Findings

Analysis of four systems shows industry convergence of operators.

Framework enables testable predictions and diagnostic benchmarks.

Addresses structural limitations of context expansion in LLMs.

Abstract

The prevailing approach to improving large language model (LLM) reasoning has centered on expanding context windows, implicitly assuming that more tokens yield better performance. However, empirical evidence - including the "lost in the middle" effect and long-distance relational degradation - demonstrates that contextual space exhibits structural gradients, salience asymmetries, and entropy accumulation under transformer architectures. We introduce Context Cartography, a formal framework for the deliberate governance of contextual space. We define a tripartite zonal model partitioning the informational universe into black fog (unobserved), gray fog (stored memory), and the visible field (active reasoning surface), and formalize seven cartographic operators - reconnaissance, selection, simplification, aggregation, projection, displacement, and layering - as transformations governing information transitions between and within zones. The operators are derived from a systematic coverage analysis of all non-trivial zone transformations and are organized by transformation type (what the operator does) and zone scope (where it applies). We ground the framework in the salience geometry of transformer attention, characterizing cartographic operators as necessary compensations for linear prefix memory, append-only state, and entropy accumulation under expanding context. An analysis of four contemporary systems (Claude Code, Letta, MemOS, and OpenViking) provides interpretive evidence that these operators are converging independently across the industry. We derive testable predictions from the framework - including operator-specific ablation hypotheses - and propose a diagnostic benchmark for empirical validation.