Multimodal Machine Unlearning Benchmarked: Oldest Methods Win, None Truly Work
_vztu · x · 2026-10-17
A team led by researchers from NTU, Columbia and Texas A&M (including Hsuan-Tien Lin) benchmarked multimodal unlearning and found the strongest methods are among the oldest and simplest — GADiff (2022) and SCRUB (2023) — while the newest method tested, MANU (2025), ranks only 6th and 5th.
Bottom line: no existing method solves real-world multimodal unlearning, and the evaluation bar needs to rise.
More from Research
- SAE features are only partially compositional, new paper finds — OpenSAE · 2026-10-17
- Contact-UAN: MIT's residual world model fixes humanoid sim-to-real with 6.5 minutes of data — kevin_zakka · 2026-10-17
- ROS 2 LiDAR localization runs 390 m outdoors on handheld Livox MID-360 — rsasaki0109 · 2026-10-17
- New paper asks how seriously to take AI welfare, with constraints from consciousness epistemology — birchlse · 2026-10-17
- GPT-6 Astra fastest-rising method on Papers with Code, 27 papers in 30 days — NielsRogge · 2026-10-17
- Embodied Astra tops RoboMME at 92.1% — beating humans by just keeping full visual history in context — hankyang94 · 2026-10-17