# AgroSense 2.0: Cross-Modal Transformer Fusion with Geospatial Raster Integration and Interpretable Multi-Task Learning for Precision Crop Recommendation

arXiv: https://arxiv.org/abs/2606.21892 · HTML: https://mlsanity.com/paper/2606.21892/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 17 references that are in the map.
- First posted on arXiv 20 Jun 2026.

## Builds on (most cited first, 6 of 6)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [Revisiting Deep Learning Models for Tabular Data](https://mlsanity.com/paper/2106.11959/) (2021)
- [Label-Efficient Learning in Agriculture: A Comprehensive Review](https://mlsanity.com/paper/2305.14691/) (2023)
- [AgroXAI: Explainable AI-Driven Crop Recommendation System for Agriculture 4.0](https://mlsanity.com/paper/2412.16196/) (2024)
- [AgroSense: An Integrated Deep Learning System for Crop Recommendation via Soil Image Analysis and Nutrient Profiling](https://mlsanity.com/paper/2509.01344/) (2025)

## Abstract (arXiv)

Crop recommendation systems in precision agriculture have long suffered from a fundamental modality gap: visual soil characterization and chemical nutrient profiling are typically treated as independent inference problems, with fusion often reduced to late-stage feature concatenation. AgroSense~2.0 addresses this limitation through three architectural advances. First, we introduce continental-scale geospatial integration via a seven-band soil raster (textttindiasoil7bands.tif) spanning India, encoding Nitrogen, pH, SOC, Clay, Sand, Silt, and Bulk Density as 32×32 spatial patches, a modality entirely absent from prior work. Second, we replace naive feature concatenation with a cross-modal Transformer fusion module, where tabular nutrient features attend over image representations via multi-head attention, enabling richer inter-modal dependency modeling than shallow fusion. Third, we adopt a multi-task objective jointly optimizing soil classification and crop recommendation through a shared backbone, improving generalization via complementary cross-task signal. To enhance interpretability, we apply TreeSHAP to the tabular branch, revealing crop-conditioned nutrient sensitivity: humidity and rainfall emerge as the most influential features globally, while crop-specific profiles diverge meaningfully rainfall dominates rice, nitrogen and potassium dominate maize, and humidity and nitrogen dominate coffee. These explanations provide transparency into model decisions and surface both agronomically consistent patterns and dataset-specific divergences worth further study. Together, these contributions establish AgroSense~2.0 as a more principled, interpretable, and geospatially grounded framework for precision agriculture.