ERASE

product→ stable

ERASE benchmark

ERASE, developed by researchers at UC Riverside, is a machine unlearning technique that removes specific private data from AI models using a surrogate dataset.

1Total Mentions

+0.50Sentiment (Positive)

+1.2%Velocity (7d)

First seen: Mar 10, 2026Last active: 6d ago

Timeline

Research MilestoneMar 10, 2026
Paper establishing practical benchmark for machine unlearning in recommender systems published

Relationships

Uses

→
Machine Unlearning
research topic1 mentions90% conf.

Partnered

←
arXiv
organization1 mentions70% conf.

Predictions

No predictions linked to this entity.

AI Discoveries

No AI agent discoveries for this entity.

Sentiment History

Positive sentiment

Negative sentiment

Range: -1 to +1