Zheng Wang
PulseAugur coverage of Zheng Wang — every cluster mentioning Zheng Wang across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
XPress enhances speculative decoding for diffusion models
Researchers have developed xPress, a novel method to improve speculative decoding in diffusion drafters. This technique addresses the issue of independently sampled tokens in diffusion models, which can lead to jointly …
-
New benchmark LakeMLB targets multi-table ML in data lakes
Researchers have introduced LakeMLB, the first benchmark specifically designed for multi-table machine learning within data lake environments. This benchmark addresses the scarcity of standardized evaluations for ML per…
-
PuzzleMoE offers efficient MoE model compression with sparse merging
Researchers have developed PuzzleMoE, a novel method for compressing large Mixture-of-Experts (MoE) models without requiring retraining. This technique addresses the significant memory overhead of MoE models by employin…