AI/ ai · machine-learning · llm · security

Fine-Tuning an Edited AI Model Wipes Most of the Edits

New research finds that knowledge edits to large language models decay sharply once fine-tuning begins, cutting both a safety lifeline and a liability.

Correcting a deployed AI model's facts is already hard — keeping those corrections through further training may be harder.

Researchers tested 254 experimental configurations to measure what happens to knowledge edits when a large language model is subsequently fine-tuned. The short answer: the edits mostly disappear. The starkest example in the study involves AlphaEdit applied to GPT-J on the zsRE benchmark, where more than a quarter of previously successful edits failed after fine-tuning. The team also found a counterintuitive wrinkle — fine-tuning only the non-edited layers caused more edit decay than fine-tuning all layers at once.

This matters in two directions that pull against each other. On one hand, a company that carefully patches an AI's bad outputs before deploying it can lose those patches the moment any downstream fine-tuning touches the model. On the other, anyone trying to scrub a malicious or covert edit out of a model has a practical tool: targeted fine-tuning of the edited layers removed edits while only modestly hurting performance elsewhere.

Knowledge editing was pitched as a cheaper alternative to full retraining — a way to correct LLMs without starting over. What this study quietly implies is that any pipeline treating edits as durable configuration rather than fragile state is building on shaky ground.

TR

The Revision

Written by an AI system from the public sources credited above. How we write →