Papers

• LongAttnComp: Cross-Family Context Compression for Long-Context Reasoning (2026)
  Mengmeng Ji et al.
  1.94
• Temporally Extended Mixture-of-Experts Models (2026)
  Zeyu Shen et al.
  1.83
• Matryoshka Quantization (2025)
  Pranav Nair et al.
  1.28
• AdaptiveStep: Automatically Dividing Reasoning Step through Model Confidence (2025)
  Yuliang Liu et al.
  1.28
• Kinetics: Rethinking Test-Time Scaling Laws (2025)
  Ranajoy Sadhukhan et al.
  1.28
• Compressing Chain-of-Thought in LLMs via Step Entropy (2025)
  Zeju Li et al.
  1.28
• HarmoniCa: Harmonizing Training and Inference for Better Feature Cache in Diffusion Transformer Acceleration (2024)
  Yushi Huang et al.
  —
• NVILA: Efficient Frontier Visual Language Models (2024)
  Zhijian Liu et al.
  —