PPaperPicks

Lujun Li

Hong Kong University of Science and Technology, Hong Kong, SAR, China

25 papers at tracked venues · 21 at CORE A* · active 20242026

Venues

Frequent coauthors

Papers

  1. BTC-LLM: Efficient Sub-1-Bit LLM Quantization via Learnable Transformation and Binary Codebook
  2. Bit-by-Bit: Progressive QAT Strategy with Outlier Channel Splitting for Stable Low-Bit LLMs
  3. Outlier Matters: Efficient Long-to-Short Reasoning via Outlier-Guided Model Merging
  4. QaRL: Rollout-Aligned Quantization-Aware RL for Fast and Stable Training under Training-Inference Mismatch
  5. Sub-MoE: Efficient Mixture-of-Expert LLMs Compression via Subspace Expert Merging
    AAAI 2026 · Lujun Li
  6. AIRA: Activation-Informed Low-Rank Adaptation for Large Models
    ICCV 2025 · Lujun Li
  7. BayesKD: Bayesian Knowledge Distillation for Compact LLMs in Constrained Fine-tuning Scenarios
  8. Can Compressed LLMs Truly Act? An Empirical Evaluation of Agentic Capabilities in LLM Compression
  9. Delta Decompression for MoE-based LLMs Compression
  10. Efficient Fine-Tuning of Large Models Via Nested Low-Rank Adaptation
    ICCV 2025 · Lujun Li
  11. MoE-SVD: Structured Mixture-of-Experts LLMs Compression via Singular Value Decomposition
  12. Outlier-Aware Model Merging for Efficient Multitask Inference
  13. ParZC: Parametric Zero-Cost Proxies for Efficient NAS
  14. STBLLM: Breaking the 1-Bit Barrier with Structured Binary LLMs
  15. Adaptive Layer Sparsity for Large Language Models via Activation Correlation Assessment
  16. AttnZero: Efficient Attention Discovery for Vision Transformers
    ECCV 2024 · Lujun Li
  17. Auto-DAS: Automated Proxy Discovery for Training-Free Distillation-Aware Architecture Search
  18. Auto-GAS: Automated Proxy Discovery for Training-Free Generative Architecture Search
    ECCV 2024 · Lujun Li
  19. Auto-Prox: Training-Free Vision Transformer Architecture Search via Automatic Proxy Discovery
  20. DetKDS: Knowledge Distillation Search for Object Detectors
    ICML 2024 · Lujun Li
  21. Discovering Sparsity Allocation for Layer-wise Pruning of Large Language Models
    NeurIPS 2024 · Lujun Li
  22. LPZero: Language Model Zero-cost Proxy Search from Zero
  23. Pruner-Zero: Evolving Symbolic Pruning Metric From Scratch for Large Language Models
  24. SasWOT: Real-Time Semantic Segmentation Architecture Search WithOut Training
  25. UniADS: Universal Architecture-Distiller Search for Distillation Gap