Footprint-Guided Exemplar-Free Continual Histopathology Report Generation

TL;DR

This paper presents a novel exemplar-free continual learning framework for histopathology report generation from whole-slide images, using compact domain footprints for generative replay to adapt to evolving clinical data without storing raw images.

Contribution

It introduces a footprint-based generative replay method that enables continual learning in histopathology report generation without exemplars or raw data storage.

Findings

Outperforms exemplar-free and rehearsal baselines on public benchmarks.

Effective in handling domain shifts and evolving reporting conventions.

Supports domain-agnostic deployment without explicit domain labels.

Abstract

Rapid progress in vision-language modeling has enabled pathology report generation from gigapixel whole-slide images, but most approaches assume static training with simultaneous access to all data. In clinical deployment, however, new organs, institutions, and reporting conventions emerge over time, and sequential fine-tuning can cause catastrophic forgetting. We introduce an exemplar-free continual learning framework for WSI-to-report generation that avoids storing raw slides or patch exemplars. The core idea is a compact domain footprint built in a frozen patch-embedding space: a small codebook of representative morphology tokens together with slide-level co-occurrence summaries and lightweight patch-count priors. These footprints support generative replay by synthesizing pseudo-WSI representations that reflect domain-specific morphological mixtures, while a teacher snapshot provides pseudo-reports to supervise the updated model without retaining past data. To address shifting reporting conventions, we distill domain-specific linguistic characteristics into a compact style descriptor and use it to steer generation. At inference, the model identifies the most compatible descriptor directly from the slide signal, enabling domain-agnostic setup without requiring explicit domain identifiers. Evaluated across multiple public continual learning benchmarks, our approach outperforms exemplar-free and limited-buffer rehearsal baselines, highlighting footprint-based generative replay as a practical solution for deployment in evolving clinical settings.