PPaperPicks

Yunhua Zhou

15 papers at tracked venues · 12 at CORE A* · active 20242026

Venues

Frequent coauthors

Papers

  1. How to Set the Learning Rate for Large-Scale Pre-training?
    ACL 2026 · Yunhua Zhou
  2. Which Reasoning Trajectories Teach Students to Reason Better? A Simple Metric of Informative Alignment
  3. BitStack: Any-Size Compression of Large Language Models in Variable Memory Environments
  4. Capability Salience Vector: Fine-grained Alignment of Loss and Capabilities for Downstream Task Scaling Law
  5. Data Mixing Laws: Optimizing Data Mixtures by Predicting Language Modeling Performance
  6. Firewall Routing: Blocking Leads to Better Hybrid Inference for LLMs
  7. Implicit Reward as the Bridge: A Unified View of SFT and DPO Connections
  8. Pre-Trained Policy Discriminators are General Reward Models
  9. Revisiting the Test-Time Scaling of o1-like Models: Do they Truly Possess Test-Time Scaling Capabilities?
  10. Towards Universality: Studying Mechanistic Similarity Across Language Model Architectures
  11. AnyGPT: Unified Multimodal LLM with Discrete Sequence Modeling
  12. Code Needs Comments: Enhancing Code LLMs with Comment Augmentation
  13. DenoSent: A Denoising Objective for Self-Supervised Sentence Representation Learning
  14. Memorize Step by Step: Efficient Long-Context Prefilling with Incremental Memory and Decremental Chunk
  15. Turn Waste into Worth: Rectifying Top-k Router of MoE