Zhouchen Lin

11 papers Peking University

• AdaMSS: Adaptive Multi-Subspace Approach for Parameter-Efficient Fine-Tuning
• Improving Model Representation and Reducing KV Cache via Skip Connections with First Value Heads
• Inverse Methods for Missing Data Imputation
• Language Ranker: A Lightweight Ranking framework for LLM Decoding
• MME-VideoOCR: Evaluating OCR-Based Capabilities of Multimodal LLMs in Video Scenarios
• On the $O(\frac{\sqrt{d}}{K^{1/4}})$ Convergence Rate of AdamW Measured by $\ell_1$ Norm
• PaZO: Preconditioned Accelerated Zeroth-Order Optimization for Fine-Tuning LLMs
• Projective Equivariant Networks via Second-order Fundamental Differential Invariants
• PseuZO: Pseudo-Zeroth-Order Algorithm for Training Deep Neural Networks
• Stepsize anything: A unified learning rate schedule for budgeted-iteration training
• Time-o1: Time-Series Forecasting Needs Transformed Label Alignment