This dataset models regression cascades caused by model updates in tool-using AI systems. It detects when update-driven drift rises, safety buffers weaken due to partial canary coverage, governance lag delays rollback, and tool coupling propagates regressions across workflows, crossing the five-node cascade threshold into an unrecoverable update regression cascade.
This dataset models a five-node cascade: four interacting instability drivers and one emergent… See the full description on the dataset page:
https://huggingface.co/datasets/ClarusC64/ai-5node-drift-buf-lag-tool-update-regression-v0.1.
