Noisy Certified Unlearning Hits Utility Ceiling in Mass Deletion
gkdziugaite · x · 2026-07-07
Existing noise-based certified machine unlearning methods have a fundamental limitation: the amount of noise required to guarantee unlearning certification severely damages model utility in large-scale deletion request scenarios, creating a utility ceiling. This bottleneck prompted researchers to develop the ALU framework, which leverages public data to reduce noise-induced costs.
Related event: ICML 2026 Paper Proposes ALU Framework for Large-Scale Machine Unlearning(5 posts)→
More from Research
- Diamandis: AI Cancer Navigator and $100 Genome Sequencing Close the Care Gap — PeterDiamandis · 2026-10-27
- EMNLP Oral: training-free Nexus Sampling cuts KV-cache memory 10x; OpenResearcher tops BrowseComp-Plus at 54.8% — yuz9yuz · 2026-10-27
- ICML 2026 position paper: LLM 'thinking traces' are a harmful anthropomorphism — rao2z · 2026-10-27
- Forking gap follows a power law in context frequency; masking rare n-grams shrinks it — ZimingLiu11 · 2026-10-27
- How forking happens: n-gram tables split logits into non-interacting subspaces — ZimingLiu11 · 2026-10-27
- The best LoRA explainer video yet: rank constraints meet intrinsic dimension — sophiamyang · 2026-10-27