Neural network-based continuous and discrete survival prediction methods, including quantile discretization and piecewise hazard models, outperform traditional approaches.

Håvard Kvamme, Ørnulf Borgan

This paper analyzes MCMC challenges in Bayesian neural networks, showing that non-converged chains can still produce accurate predictive distributions.

Theodore Papamarkou, Jacob Hinkle, M. Todd Young et al.

Proposes Dual-path RNN (DPRNN) for efficient long sequence modeling in single-channel speech separation, achieving 20x smaller models with state-of-the-art performance.

Yi Luo, Zhuo Chen, Takuya Yoshioka

This study analyzes self-play vs. human model learning in cooperation, proposing methods to improve human-AI collaboration using Overcooked experiments.

Micah Carroll, Rohin Shah, Mark K. Ho et al.

Proposes a GCN-GRU deep learning model for dynamic uncertain opinion prediction, addressing scalability, heterogeneity, and conflicting evidence.

Xujiang Zhao, Feng Chen, Jin-Hee Cho

Proposes Fi-GNN, a graph neural network model that explicitly models feature interactions, improving CTR prediction by 4.2% AUC on Criteo.

Zekun Li, Zeyu Cui, Shu Wu et al.

Proposes Meta Neural Trojan Detection (MNTD) using jumbo learning, achieving 97% AUC in black-box Trojan detection across diverse datasets.

Xiaojun Xu, Qi Wang, Huichen Li et al.

DexPilot enables low-cost vision-based teleoperation of a 23 DoA robotic system by observing the bare human hand.

Ankur Handa, Karl Van Wyk, Wei Yang et al.

Combines RNN with low-rank Gaussian Copula process for high-dimensional multivariate forecasting, achieving significant accuracy improvements over state-of-the-art methods.

David Salinas, Michael Bohlke-Schneider, Laurent Callot et al.

Checkmate breaks the memory wall with optimal tensor rematerialization, enabling 5.1x larger inputs.

Paras Jain, Ajay Jain, Aniruddha Nrusimha et al.

Proposes a 3D scene graph framework combining deep learning detection and multi-view consistency for comprehensive semantic scene understanding.

Iro Armeni, Zhi-Yang He, JunYoung Gwak et al.

ZeRO partitions model states to enable trillion-parameter training with reduced memory and high efficiency.

Samyam Rajbhandari, Jeff Rasley, Olatunji Ruwase et al.

Attention-based CNN improves face forgery detection, achieving 99.76% AUC and better localization.

Hao Dang, Feng Liu, Joel Stehouwer et al.

Proposes DistilBERT, a 40% smaller model with 97% performance, 60% faster inference, via pre-training distillation with triple loss.

Victor Sanh, Lysandre Debut, Julien Chaumond et al.

Selective-Backprop accelerates deep learning by prioritizing high-loss samples, achieving up to 3.5x faster convergence.

Angela H. Jiang, Daniel L. -K. Wong, Giulio Zhou et al.

TagSLAM integrates AprilTags with factor graph optimization, enabling robust monocular SLAM with high accuracy and real-time performance.

Bernd Pfrommer, Kostas Daniilidis

Bayesian optimization reveals that low-connectivity (even zero spectral radius) Reservoir networks can effectively forecast chaotic systems.

Aaron Griffith, Andrew Pomerance, Daniel J. Gauthier

BillSum dataset for US legislative text summarization shows cross-state applicability.

Anastassia Kornilova, Vlad Eidelman

DiffTaichi is a differentiable programming language for physical simulation, achieving significant performance improvements.

Yuanming Hu, Luke Anderson, Tzu-Mao Li et al.

FaSNet, a time-domain neural beamformer, outperforms traditional methods with 14.3% WER reduction on CHiME-3.

Yi Luo, Enea Ceolini, Cong Han et al.