Hao Gu

Optimizing Temporal and Spatial Efficiency for Chain-of-Thought Reasoning in Large Language Models

Drop or Merge? Hybrid MoE LLMs Compressors via Metric-Driven Adaptive Allocation

BTC-LLM: Efficient Sub-1-Bit LLM Quantization via Learnable Transformation and Binary Codebook

Light-Search: Reducing Retrieval Cost in RAG via Curriculum-Based Policy Optimization

Bit-by-Bit: Progressive QAT with Outlier Channel Splitting for Stable Low-Bit LLMs

Delta Decompression for MoE-based LLMs Compression

MoE-SVD: Structured Mixture-of-Experts LLMs Compression via Singular Value Decomposition