# Spatial Atlas: Compute-Grounded Reasoning for Spatial-Aware Research Agent Benchmarks

arXiv: https://arxiv.org/abs/2604.12102 · HTML: https://mlsanity.com/paper/2604.12102/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 24 of its 49 references that are in the map.
- First posted on arXiv 13 Apr 2026.

## Placed in

- LLM reasoning / [Spatial and embodied reasoning](https://mlsanity.com/lr/spatial-and-embodied-reasoning/)
- LLM agents and tool use / [Multimodal and embodied agents](https://mlsanity.com/agents/multimodal-and-embodied-agents/)

## Builds on (most cited first, 24 of 24)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations](https://mlsanity.com/paper/1602.07332/) (2016)
- [A Survey on Large Language Model based Autonomous Agents](https://mlsanity.com/paper/2308.11432/) (2023)
- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [Evaluating Object Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2305.10355/) (2023)
- [OpenHands: An Open Platform for AI Software Developers as Generalist Agents](https://mlsanity.com/paper/2407.16741/) (2024)
- [SAM 3: Segment Anything with Concepts](https://mlsanity.com/paper/2511.16719/) (2025)
- [SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities](https://mlsanity.com/paper/2401.12168/) (2024)
- [Depth Anything 3: Recovering the Visual Space from Any Views](https://mlsanity.com/paper/2511.10647/) (2025)
- [MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering](https://mlsanity.com/paper/2410.07095/) (2024)
- [Scene Graph Generation by Iterative Message Passing](https://mlsanity.com/paper/1701.02426/) (2017)
- [Florence-2: Advancing a Unified Representation for a Variety of Vision Tasks](https://mlsanity.com/paper/2311.06242/) (2023)
- [AutoGluon-Tabular: Robust and Accurate AutoML for Structured Data](https://mlsanity.com/paper/2003.06505/) (2020)
- [Auto-Sklearn 2.0: Hands-free AutoML via Meta-Learning](https://mlsanity.com/paper/2007.04074/) (2020)
- [Active Prompting with Chain-of-Thought for Large Language Models](https://mlsanity.com/paper/2302.12246/) (2023)
- [Large Language Models for Automated Data Science: Introducing CAAFE for Context-Aware Automated Feature Engineering](https://mlsanity.com/paper/2305.03403/) (2023)
- [Reasoning Paths with Reference Objects Elicit Quantitative Spatial Reasoning in Large Vision-Language Models](https://mlsanity.com/paper/2409.09788/) (2024)
- [Scene Graph Reasoning for Visual Question Answering](https://mlsanity.com/paper/2007.01072/) (2020)
- [Towards a Tighter Bound on Possible-Rendezvous Areas: Preliminary Results](https://mlsanity.com/paper/2206.12756/) (2022)
- [Spatial Computing Opportunities in Biomedical Decision Support: The Atlas-EHR Vision](https://mlsanity.com/paper/2305.09675/) (2023)
- [FieldWorkArena: Agentic AI Benchmark for Real Field Work Tasks](https://mlsanity.com/paper/2505.19662/) (2025)
- [Towards Spatially-Lucid AI Classification in Non-Euclidean Space: An Application for MxIF Oncology Data](https://mlsanity.com/paper/2402.14974/) (2024)

## Abstract (arXiv)

We describe compute-grounded reasoning (CGR), a design pattern in which code computes selected sub-problems from explicit intermediate representations before a language model answers. Spatial Atlas implements CGR as an Agent2Agent (A2A) server with a spatial question-answering handler and a machine-learning engineering handler. The spatial handler asks a language model to extract a scene graph, and code then fills in missing distances and checks the extracted safety rules. A separate benchmark driver can also run a strict metric bridge. It computes the gap for horizontal-gap questions from segmentation masks and a reconstructed point map, and it passes that gap to the answering model as a fact. The bridge returns a fixed unavailable answer when an evidence check fails, and it never falls back to model-estimated coordinates. The ML-engineering handler generates pipeline code, parses validation scores, and caps the number of repair and refinement passes. Its code execution is off by default. The repository also provides four run modes that can write label-free journals, a shuffled-image control mapping, and journal validators that reject label-bearing fields. We report one private label-free operational run in which four paths each wrote eight prediction rows with zero retries. Labels stayed sealed, and no score was computed, so this run establishes operational integrity only. We report no FieldWorkArena result because the benchmark data were not accessible. We also omit every performance, latency, and resource-use number that lacks a reproducible run artifact.