MOAI speeds non-interactive homomorphic transformer inference
It cuts BERT-base evaluation time 52.8% versus THOR and reports 2.36 minutes amortized per input on one GPU.
Stories per day · 30 days
2 RSS
Relatedaustria1ecg1foundation-model1homomorphic-encryption1llama1medical-ai1nlp1party-classification1political-speech1privacy1secure-inference1transformers1vq-vae1
6 stories, newest first · one card per story
Stories tagged #bert
QECGBERT tokenizes ECG waveforms for BERT-style pretraining
It is pretrained on more than 800,000 unlabeled MIMIC-IV ECGs and fine-tuned on three labeled datasets.
Fine-tuned BERT predicts Austrian MPs' parties from speeches
Governing parties grow more alike in speech, opposition diverges, and switchers adapt only after changing party.
QCATS: Query Context-Aware Transformer Slicing for Efficient Predictive Query Processing
QCATS slices transformer FFNs using query predicates, cutting in-database inference latency up to 4.42x.
OPFL: Optimistic Verification of Federated Learning via Empirical Boundary
OPFL verifies federated learning inside MPC with an empirical gradient boundary, stopping model poisoning at much lower cost.
Microsoft Finds Ransomware Group Using Same Attack Blueprint Across Multiple Malware Families
Microsoft says affiliate Storm-2570 reused one playbook across Qilin, DragonForce, Anubis, and BERT.
2