# Constructing coherent spatial memory in LLM agents through graph rectification

arXiv: https://arxiv.org/abs/2510.04195 · HTML: https://mlsanity.com/paper/2510.04195/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Agent memory systems, one of 16 directions in LLM agents and tool use.
- Builds on 20 of its 36 references that are in the map.
- First posted on arXiv 5 Oct 2025.

## Placed in

- LLM agents and tool use / [Agent memory systems](https://mlsanity.com/agents/agent-memory-systems/)

## Builds on (most cited first, 20 of 20)

- [ORB-SLAM: a Versatile and Accurate Monocular SLAM System](https://mlsanity.com/paper/1502.00956/) (2015)
- [Logic-LM: Empowering Large Language Models with Symbolic Solvers for Faithful Logical Reasoning](https://mlsanity.com/paper/2305.12295/) (2023)
- [ConceptGraphs: Open-Vocabulary 3D Scene Graphs for Perception and Planning](https://mlsanity.com/paper/2309.16650/) (2023)
- [TextWorld: A Learning Environment for Text-based Games](https://mlsanity.com/paper/1806.11532/) (2018)
- [Interactive Fiction Games: A Colossal Adventure](https://mlsanity.com/paper/1909.05398/) (2019)
- [LINC: A Neurosymbolic Approach for Logical Reasoning by Combining Language Models with First-Order Logic Provers](https://mlsanity.com/paper/2310.15164/) (2023)
- [SceneGraphFusion: Incremental 3D Scene Graph Prediction from RGB-D Sequences](https://mlsanity.com/paper/2103.14898/) (2021)
- [If beam search is the answer, what was the question?](https://mlsanity.com/paper/2010.02650/) (2020)
- [SG-Nav: Online 3D Scene Graph Prompting for LLM-based Zero-shot Object Navigation](https://mlsanity.com/paper/2410.08189/) (2024)
- [SatLM: Satisfiability-Aided Language Models Using Declarative Prompting](https://mlsanity.com/paper/2305.09656/) (2023)
- [Mind's Eye of LLMs: Visualization-of-Thought Elicits Spatial Reasoning in Large Language Models](https://mlsanity.com/paper/2404.03622/) (2024)
- [AlphaMaze: Enhancing Large Language Models' Spatial Intelligence via GRPO](https://mlsanity.com/paper/2502.14669/) (2025)
- [KARMA: Leveraging Multi-Agent LLMs for Automated Knowledge Graph Enrichment](https://mlsanity.com/paper/2502.06472/) (2025)
- [OneEdit: A Neural-Symbolic Collaboratively Knowledge Editing System](https://mlsanity.com/paper/2409.07497/) (2024)
- [MANGO: A Benchmark for Evaluating Mapping and Navigation Abilities of Large Language Models](https://mlsanity.com/paper/2403.19913/) (2024)
- [Efficiently Closing Loops in LiDAR-Based SLAM Using Point Cloud Density Maps](https://mlsanity.com/paper/2501.07399/) (2025)
- [Research on Navigation Methods Based on LLMs](https://mlsanity.com/paper/2504.15600/) (2025)
- [Scene Graph Modification as Incremental Structure Expanding](https://mlsanity.com/paper/2209.09093/) (2022)
- [ConVer-G: Concurrent versioning of knowledge graphs](https://mlsanity.com/paper/2409.04499/) (2024)
- [Can LLMs Learn to Map the World from Local Descriptions?](https://mlsanity.com/paper/2505.20874/) (2025)

## Built on by (most cited first, 2 of 2)

- [From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms](https://mlsanity.com/paper/2605.06716/) (2026)
- [Reasmory: 3D Reconstruction as Explicit Memory for VLMs Spatial Reasoning](https://mlsanity.com/paper/2606.00963/) (2026)

## Abstract (arXiv)

Given a map description through global traversal navigation instructions, an LLM can often infer the implicit spatial layout and answer user queries by providing shortest paths. However, such context-dependent querying becomes incapable as environments grow larger, motivating the need for incremental map construction that builds a complete topological graph from stepwise observations. We propose LLM-MapRepair, a framework for LLM-driven construction and map repair, designed to detect, localize, and correct structural inconsistencies in incrementally constructed navigation graphs. Our contributions include a Version Control mechanism for graph construction, an Edge Impact Score for repair prioritization, and a cleaned variant of the MANGO benchmark tailored for LLM-driven map construction and repair. We evaluate the framework on four evaluation settings: a synthetic per-component ablation (gpt-4.1, n=20 seeds per cell), a cross-vendor sweep over seven LLMs from OpenAI, Anthropic, and Google on both synthetic and TextWorld procedurally-generated text-adventure games, a repair-stage evaluation on all 42 cleaned-MANGO games with non-zero residual conflicts (534 conflicts; three vendors x three modes plus two non-LLM references), and an end-to-end natural-text deployment on Chapters 16-17 of Dream of the Red Chamber. On the DRC deployment, LLM-MapRepair achieves 94.3% node recall (+8.6 pp over direct LLM mapping) and 88.2% edge recall (+55.8 pp), using GPT-4.1; the recall improvements come with predicted node and edge counts that are roughly 4x the ground-truth counts (Table 4), reflecting the discretization-driven over-generation trade-off we discuss in the Limitations.