language tasks
- Accurate and Efficient Low-Rank Model Merging in Core Space
- All You Need is One: Capsule Prompt Tuning with a Single Vector
- Differentiable Sparsity via $D$-Gating: Simple and Versatile Structured Penalization
- List-Level Distribution Coupling with Applications to Speculative Decoding and Lossy Compression
- Stepsize anything: A unified learning rate schedule for budgeted-iteration training
- Understanding and Enhancing Mask-Based Pretraining towards Universal Representations