Yutao Zeng

Yutao Zeng on Hugging Face Daily Papers: 13 papers, 4 in the top 3 of their day, 431 upvotes.

  1. Mixture-of-Depths Attention 77 upvotes, #7 of 2026-03-17
  2. Virtual Width Networks 34 upvotes, #3 of 2025-11-17
  3. UltraMemV2: Memory Networks Scaling to 120B Parameters with Superior Long-Context Learning 34 upvotes, #6 of 2025-08-27
  4. Stepsize anything: A unified learning rate schedule for budgeted-iteration training 5 upvotes, #35 of 2025-06-03
  5. Scaling Law for Quantization-Aware Training 67 upvotes, #3 of 2025-05-22
  6. Efficient Pretraining Length Scaling 19 upvotes, #10 of 2025-04-23
  7. Frac-Connections: Fractional Extension of Hyper-Connections 19 upvotes, #10 of 2025-03-19
  8. HybridNorm: Towards Stable and Efficient Transformer Training via Hybrid Normalization 17 upvotes, #10 of 2025-03-07
  9. Scale-Distribution Decoupling: Enabling Stable and Effective Training of Large Language Models 13 upvotes, #10 of 2025-02-26
  10. Over-Tokenized Transformer: Vocabulary is Generally Worth Scaling 22 upvotes, #3 of 2025-01-29
  11. Ultra-Sparse Memory Network 19 upvotes, #7 of 2024-11-22
  12. Polynomial Composition Activations: Unleashing the Dynamics of Large Language Models 20 upvotes, #3 of 2024-11-07
  13. Hyper-Connections 17 upvotes, #5 of 2024-10-01

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.