LoTUS: New Machine Unlearning Method Removes Training Data Influence Without Full Retraining
Researchers have introduced LoTUS, a machine unlearning technique that removes the influence of specific training samples from pre-trained AI models without requiring costly retraining from scratch. The method works by smoothing prediction probabilities up to an information-theoretic bound, reducing overconfidence caused by data memorization. Accepted at CVPR 2025, LoTUS addresses a growing need for practical data removal in large-scale models where full retraining is infeasible.
LoTUS (Large-Scale Machine Unlearning with a Taste of Uncertainty) is a novel method designed to selectively eliminate the influence of training samples from pre-trained models, a capability increasingly important for privacy compliance and data governance. The approach smooths a model's prediction probabilities up to an information-theoretic bound, counteracting the overconfidence that arises when models memorize specific training data. The authors evaluated LoTUS on both Transformer and ResNet18 architectures across five public datasets, benchmarking it against eight competing baselines. Notably, the evaluation extends to ImageNet1k, a large-scale dataset where retraining from scratch is computationally impractical, better simulating real-world deployment conditions. To support evaluation without a retrained reference model, the researchers also introduce a new metric called the Retrain-Free Jensen-Shannon Divergence (RF-JSD). Experimental results indicate LoTUS outperforms current state-of-the-art methods in both efficiency and effectiveness. The paper has been accepted as a main conference paper at CVPR 2025.
What's missing
The generalizability of RF-JSD as a community-adopted metric remains unvalidated by independent studies.
What different sources said
- arXiv cs.AICenter
Routing-Aware Expert Calibration for Machine Unlearning in Mixture-of-Experts Language Models
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.