Zekun Wang

4 papers Harbin Institute of Technology

• Deep Taxonomic Networks for Unsupervised Hierarchical Prototype Discovery
• Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free
• Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free
• Scaling Computer-Use Grounding via User Interface Decomposition and Synthesis