All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.
- Synchronized the repository documentation, model card, and citation metadata with the ECCV 2026 camera-ready paper.
- Expanded the Notes from the Authors with discussions of generalization, case aggregation, benchmarking, and Stage 2 hyperparameters.
- Updated the reported evaluation results and supporting analysis to cover all sixteen held-out tasks.
- Two-stage training pipeline (self-supervised slide encoding + supervised case-level alignment)
- CLI interface (
moozy train stage1,moozy train stage2,moozy encode) - HuggingFace Hub integration for checkpoint downloads
- Multi-task supervised training with classification and survival objectives
- Slurm job templates for single-GPU, multi-GPU, and multi-node training
- CI workflow with ruff, mypy, and compile checks
- Training documentation for Stage 1, Stage 2, and encoding