OmniTabBench: Large-Scale Benchmark for Tabular ML Models
Discover OmniTabBench, the largest benchmark with 3030 datasets evaluating GBDTs, neural networks, and foundation models for tabular data at scale.
SkillTrojan Backdoor Attacks on AI Skill-Based Agents
Discover how SkillTrojan exploits backdoor attacks in skill-based AI agents, revealing critical security risks and the need for advanced defenses.
MoBiE: Fast, Efficient Mixture of Binary Experts Inference
Discover MoBiE, a breakthrough in efficient inference for Mixture-of-Experts models using post-training quantization to boost speed and reduce memory.
Instance-Adaptive VAE for Efficient Variational Inference
Discover how instance-adaptive parametrization improves variational autoencoders, reducing the amortization gap for better inference accuracy and efficienc...
FedDAP: Domain-Aware Prototype Learning for Federated AI
Discover FedDAP, a domain-aware prototype learning method enhancing federated learning performance under domain shifts with dual alignment strategy.