DHA: Learning Decoupled-Head Attention from Transformer Checkpoints via Adaptive Heads Fusionby Yilong Chen, Linhao Zhang,…
LLM Dataset Inference: Did you train on my dataset?by Pratyush Maini, Hengrui Jia, Nicolas Papernot,…
Foundation Inference Models for Markov Jump Processesby David Berghaus, Kostadin Cvejoski, Patrick Seifner, Cesar Ojeda,…
Decoupled Marked Temporal Point Process using Neural Ordinary Differential Equationsby Yujee Song, Donghyun Lee, Rui…
Efficient Neural Compression with Inference-time Decodingby C. Metz, O. Bichler, A. DupretFirst submitted to arxiv…
PowerInfer-2: Fast Large Language Model Inference on a Smartphoneby Zhenliang Xue, Yixin Song, Zeyu Mi,…
Turbo Sparse: Achieving LLM SOTA Performance with Minimal Activated Parametersby Yixin Song, Haotong Xie, Zhengyan…
GraphStorm: all-in-one graph machine learning framework for industry applicationsby Da Zheng, Xiang Song, Qi Zhu,…
PSBD: Prediction Shift Uncertainty Unlocks Backdoor Detectionby Wei Li, Pin-Yu Chen, Sijia Liu, Ren WangFirst…
Scaling Graph Convolutions for Mobile Visionby William Avery, Mustafa Munir, Radu MarculescuFirst submitted to arxiv…