# From Pixels to Policy: A Multi-Agent System for Intervention and Geo-Spatial Decision Support

arXiv: https://arxiv.org/abs/2610.01870 · HTML: https://mlsanity.com/paper/2610.01870/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 56 references that are in the map.
- First posted on arXiv 1 Oct 2026.

## Builds on (most cited first, 26 of 26)

- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [OpenAI GPT-5 System Card](https://mlsanity.com/paper/2601.03267/) (2025)
- [GeoChat: Grounded Large Vision-Language Model for Remote Sensing](https://mlsanity.com/paper/2311.15826/) (2023)
- [DeepGlobe 2018: A Challenge to Parse the Earth through Satellite Images](https://mlsanity.com/paper/1805.06561/) (2018)
- [SkyEyeGPT: Unifying Remote Sensing Vision-Language Tasks via Instruction Tuning with Large Language Model](https://mlsanity.com/paper/2401.09712/) (2024)
- [OpenEarthMap: A Benchmark Dataset for Global High-Resolution Land Cover Mapping](https://mlsanity.com/paper/2210.10732/) (2022)
- [Deep Learning the City : Quantifying Urban Perception At A Global Scale](https://mlsanity.com/paper/1608.01769/) (2016)
- [A Multimodal Automated Interpretability Agent](https://mlsanity.com/paper/2404.14394/) (2024)
- [RS-Agent: Automating Remote Sensing Tasks through Intelligent Agent](https://mlsanity.com/paper/2406.07089/) (2024)
- [LANCE: Stress-testing Visual Models by Generating Language-guided Counterfactual Images](https://mlsanity.com/paper/2305.19164/) (2023)
- [ThinkGeo: Evaluating Tool-Augmented Agents for Remote Sensing Tasks](https://mlsanity.com/paper/2505.23752/) (2025)
- [FIND: A Function Description Benchmark for Evaluating Interpretability Methods](https://mlsanity.com/paper/2309.03886/) (2023)
- [GeoSynth: Contextually-Aware High-Resolution Satellite Image Synthesis](https://mlsanity.com/paper/2404.06637/) (2024)
- [Latent Diffusion Counterfactual Explanations](https://mlsanity.com/paper/2310.06668/) (2023)
- [Towards LLM Agents for Earth Observation](https://mlsanity.com/paper/2504.12110/) (2025)
- [LISAT: Language-Instructed Segmentation Assistant for Satellite Imagery](https://mlsanity.com/paper/2505.02829/) (2025)
- [Are Safer Looking Neighborhoods More Lively? A Multimodal Investigation into Urban Life](https://mlsanity.com/paper/1608.00462/) (2016)
- [A citizen science toolkit to collect human perceptions of urban environments using open street view images](https://mlsanity.com/paper/2403.00174/) (2024)
- [VectorSynth: Fine-Grained Satellite Image Synthesis with Structured Semantics](https://mlsanity.com/paper/2511.07744/) (2025)
- [Deep Learning Based Large-Scale Automatic Satellite Crosswalk Classification](https://mlsanity.com/paper/1706.09302/) (2017)
- [URSimulator: Human-Perception-Driven Prompt Tuning for Enhanced Virtual Urban Renewal via Diffusion Models](https://mlsanity.com/paper/2409.14589/) (2024)
- [Beta Distribution Learning for Reliable Roadway Crash Risk Assessment](https://mlsanity.com/paper/2511.04886/) (2025)
- [The Urban Model Platform: A Public Backbone for Modeling and Simulation in Urban Digital Twins](https://mlsanity.com/paper/2506.10964/) (2025)

## Abstract (arXiv)

Urban environments are shaped by design choices with long-term implications for health, safety, and quality of life, yet evaluating proposed interventions remains costly, time-consuming, and often impractical. Existing geospatial vision methods largely focus on monitoring urban indicators from aerial and street-view imagery, rather than proposing interventions and estimating their effects on such indicators. Moving beyond recognition, we introduce the problem of discovering interventions that improve target indicators for a given aerial or street-view image. We argue that a black-box indicator model, combined with a generative editing model, can serve as an implicit digital twin for testing intervention hypotheses. We present VIDA-Geo , a multi-agent system that explores this intervention space by coordinating segmentation, diffusion-based inpainting, and indicator scoring models to produce interventions that are both perceptually realistic and aligned with real-world policies. We evaluate our system on 8 indicators across aerial and street-view imagery, measuring changes in factors such as perceived safety and greenery. Our approach outperforms existing baselines in many cases, achieving up to 2X higher perceptual quality and policy alignment scores. Finally, our model provides users with multiple candidate interventions, supporting an expert city-planner-in-the-loop workflow.