# Semantic Context-aware mOdality fUsion Transformer (SCOUT): A Context-Aware Multimodal Transformer for Concept-Grounded Pathology Report Generation

arXiv: https://arxiv.org/abs/2605.01144 · HTML: https://mlsanity.com/paper/2605.01144/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 36 references that are in the map.
- First posted on arXiv 1 May 2026.

## Builds on (most cited first, 21 of 21)

- [SGDR: Stochastic Gradient Descent with Warm Restarts](https://mlsanity.com/paper/1608.03983/) (2016)
- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [Show and Tell: A Neural Image Caption Generator](https://mlsanity.com/paper/1411.4555/) (2014)
- [Generating Radiology Reports via Memory-driven Transformer](https://mlsanity.com/paper/2010.16056/) (2020)
- [Cross-modal Memory Networks for Radiology Report Generation](https://mlsanity.com/paper/2204.13258/) (2022)
- [Multimodal Whole Slide Foundation Model for Pathology](https://mlsanity.com/paper/2411.19666/) (2024)
- [Visual Language Pretrained Multiple Instance Zero-Shot Transfer for Histopathology Images](https://mlsanity.com/paper/2306.07831/) (2023)
- [Improving Chest X-Ray Report Generation by Leveraging Warm Starting](https://mlsanity.com/paper/2201.09405/) (2022)
- [Accelerating Data Processing and Benchmarking of AI Models for Pathology](https://mlsanity.com/paper/2502.06750/) (2025)
- [Multiple Instance Captioning: Learning Representations from Histopathology Textbooks and Articles](https://mlsanity.com/paper/2103.05121/) (2021)
- [A Generalized Deep Learning Framework for Whole-Slide Image Segmentation and Analysis](https://mlsanity.com/paper/2001.00258/) (2020)
- [HistGen: Histopathology Report Generation via Local-Global Feature Encoding and Cross-modal Context Interaction](https://mlsanity.com/paper/2403.05396/) (2024)
- [WsiCaption: Multiple Instance Generation of Pathology Reports for Gigapixel Whole-Slide Images](https://mlsanity.com/paper/2311.16480/) (2023)
- [SI-MIL: Taming Deep MIL for Self-Interpretability in Gigapixel Histopathology](https://mlsanity.com/paper/2312.15010/) (2023)
- [Activating Associative Disease-Aware Vision Token Memory for LLM-Based X-ray Report Generation](https://mlsanity.com/paper/2501.03458/) (2025)
- [Automatic Report Generation for Histopathology images using pre-trained Vision Transformers](https://mlsanity.com/paper/2311.06176/) (2023)
- [GECKO: Gigapixel Vision-Concept Contrastive Pretraining in Histopathology](https://mlsanity.com/paper/2504.01009/) (2025)
- [Historical Report Guided Bi-modal Concurrent Learning for Pathology Report Generation](https://mlsanity.com/paper/2506.18658/) (2025)
- [Pathology Report Generation and Multimodal Representation Learning for Cutaneous Melanocytic Lesions](https://mlsanity.com/paper/2502.19293/) (2025)
- [LLM-driven Medical Report Generation via Communication-efficient Heterogeneous Federated Learning](https://mlsanity.com/paper/2506.17562/) (2025)
- [S2D-ALIGN: Shallow-to-Deep Auxiliary Learning for Anatomically-Grounded Radiology Report Generation](https://mlsanity.com/paper/2511.11066/) (2025)

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

- [PathReportEval: A Systematic Benchmark for Pathology Report Generation](https://mlsanity.com/paper/2607.18448/) (2026)

## Abstract (arXiv)

Whole-slide pathology report generation requires models to integrate localized histomorphology, global tissue context, and diagnostically relevant semantic information, yet existing approaches typically rely on fixed pretrained visual representations and may fail to represent key diagnostic concepts. Here we present SCOUT: Semantic Context-aware mOdality fUsion Transformer, a concept-grounded multimodal framework that integrates local histological patterns, whole-slide context, and expert-curated diagnostic concepts. SCOUT maintains an evolving visual representation together with recursively updated slide- and concept-conditioned representations, allowing visual and contextual information to iteratively co-evolve across encoder depth. During decoding, separate attention pathways attend to these complementary streams before an adaptive multimodal fusion module combines them for each generated token. We evaluated SCOUT on TCGA-BRCA, HistAI, and REG-2025, comprising more than 20,900 WSI-report pairs across more than seven cancer types. Using common CONCHv1.5 features and evaluation protocols, SCOUT outperformed WSI-Caption, HistGen, and Bi-Gen, improving BLEU scores by 9.6% on average, METEOR by 11.0%, and ROUGE-L by 3.6% relative to the strongest competing method. On REG-2025, SCOUT additionally improved the Clinical Report Quality Score by 5.6%, indicating better preservation of clinically relevant report attributes beyond lexical similarity. Ablations showed complementary contributions from iterative context refinement and adaptive fusion, while learned gates provided inspectable token-level estimates of modality use. Our results suggest that progressive contextual conditioning is effective across heterogeneous pathology report generation settings and provides a flexible framework for integrating domain knowledge into vision-language models for computational pathology.