Keeping large language models factually up-to-date is crucial for deployment, yet costly retraining remains a challenge. Knowledge editing offers a promising alternative, but methods are only tested on small-scale or synthetic edit benchmarks. In this work, we aim to bridge research into lifelong knowledge editing to real-world edits at practically relevant scale. We first introduce WikiBigEdit; a large-scale benchmark of real-world Wikidata edits, built to automatically extend lifelong for future-proof benchmarking. In its first instance, it includes over 500K question-answer pairs for knowledge editing alongside a comprehensive evaluation pipeline. Finally, we use WikiBigEdit to study existing knowledge editing techniques' ability to incorporate large volumes of real-world facts and contrast their capabilities to generic modification techniques such as retrieval augmentation and continual finetuning to acquire a complete picture of the practical extent of current lifelong knowledge editing.
This repository contains two main components:
Contains the WikiBigEdit extraction pipeline for generating the benchmark dataset. It automatically extracts, formats, and evaluates real-world factual updates from Wikidata.
For setup and usage instructions, see the README.
Contains the adapted EasyEdit codebase for running lifelong knowledge editing experiments on WikiBigEdit. It supports structured experiment configurations via Hydra and logs results with Weights & Biases (wandb).
For details on running experiments, see the README.