# Topo-RAG: Topology-aware retrieval for hybrid text-table documents

arXiv: https://arxiv.org/abs/2601.10215 · HTML: https://mlsanity.com/paper/2601.10215/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 8 references that are in the map.
- First posted on arXiv 15 Jan 2026.

## Builds on (most cited first, 8 of 8)

- [On the Theoretical Limitations of Embedding-Based Retrieval](https://mlsanity.com/paper/2508.21038/) (2025)
- [Pneuma: Leveraging LLMs for Tabular Data Representation and Retrieval in an End-to-End System](https://mlsanity.com/paper/2504.09207/) (2025)
- [Benchmarking Deep Search over Heterogeneous Enterprise Data](https://mlsanity.com/paper/2506.23139/) (2025)
- [WARP: An Efficient Engine for Multi-Vector Retrieval](https://mlsanity.com/paper/2501.17788/) (2025)
- [CRISP: Clustering Multi-Vector Representations for Denoising and Pruning](https://mlsanity.com/paper/2505.11471/) (2025)
- [REaR: Retrieve, Expand and Refine for Effective Multitable Retrieval](https://mlsanity.com/paper/2511.00805/) (2025)
- [Bridging Queries and Tables through Entities in Table Retrieval](https://mlsanity.com/paper/2504.06551/) (2025)
- [TabRAG: Improving Tabular Document Question Answering for Retrieval Augmented Generation via Structured Representations](https://mlsanity.com/paper/2511.06582/) (2025)

## Built on by (most cited first, 1 of 1)

- [When 2D Tasks Meet 1D Serialization: On Serialization Friction in Structured Tasks](https://mlsanity.com/paper/2604.27272/) (2026)

## Abstract (arXiv)

In enterprise datasets, documents are rarely pure. They are not just text, nor just numbers; they are a complex amalgam of narrative and structure. Current Retrieval-Augmented Generation (RAG) systems have attempted to address this complexity with a blunt tool: linearization. We convert rich, multidimensional tables into simple Markdown-style text strings, hoping that an embedding model will capture the geometry of a spreadsheet in a single vector. But it has already been shown that this is mathematically insufficient. This work presents Topo-RAG, a framework that challenges the assumption that "everything is text". We propose a dual architecture that respects the topology of the data: we route fluid narrative through traditional dense retrievers, while tabular structures are processed by a Cell-Aware Late Interaction mechanism, preserving their spatial relationships. Evaluated on SEC-25, a synthetic enterprise corpus that mimics real-world complexity, Topo-RAG demonstrates an 18.4% improvement in nDCG@10 on hybrid queries compared to standard linearization approaches. It's not just about searching better; it's about understanding the shape of information.