🏢 Technion

Zero-to-Hero enhances zero-shot novel view synthesis by cleverly filtering attention maps during inference, achieving significantly higher fidelity and realism without retraining.

Optimal fault-tolerant asynchronous machine learning is achieved via a novel weighted robust aggregation framework, ensuring efficient training despite Byzantine failures and heterogeneous resources.

SLowcal-SGD, a new local update method for distributed learning, provably outperforms Minibatch-SGD and Local-SGD in heterogeneous settings by using a slow querying technique, mitigating bias from loc…

Sequential Signal Mixing Aggregation (SSMA) boosts message-passing graph neural network performance by effectively mixing neighbor features, achieving state-of-the-art results across various benchmark…

Deep learning’s generalization ability defies conventional wisdom; this paper proves that overfitting in deep neural networks is ’tempered’, neither catastrophic nor perfectly benign, for both minimal…

Hierarchical Selective Classification (HSC) improves deep learning model reliability for risk-sensitive tasks by leveraging hierarchical class relationships to provide more informative predictions eve…

This paper introduces a novel post-processing technique that significantly boosts the perceptual quality of images generated by consistency models using a joint classifier-discriminator adversarially …