Yutao Zeng
5
Papers
52
Total Citations
1
Affiliations
Affiliations
Bytedance
Papers (5)
Over-Tokenized Transformer: Vocabulary is Generally Worth Scaling
ICML 2025
26
citations
Hyper-Connections
ICLR 2025
20
citations
Polynomial Composition Activations: Unleashing the Dynamics of Large Language Models
ICLR 2025arXiv
5
citations
Stepsize anything: A unified learning rate schedule for budgeted-iteration training
NeurIPS 2025
1
citations
SimpleVQA: Multimodal Factuality Evaluation for Multimodal Large Language Models
ICCV 2025
0
citations