Publications / 论文

You can also find my articles on my Google Scholar profile.

* Equal contribution.

COLM VoidPadding: Let [VOID] Handle Padding in Masked Diffusion Language Models so that [EOS] Can Focus on Semantic Termination paper thumbnail

VoidPadding: Let [VOID] Handle Padding in Masked Diffusion Language Models so that [EOS] Can Focus on Semantic Termination

Chunyu Liu, Zhengyang Fan, Kaisen Yang, Alex Lamb
COLM 2026 Workshop on Efficient Reasoning (Spotlight), 2026
ArXiv JustQuant: You Don't Need Smoothing, SVD, or Rotation for 4-Bit Activation Quantization paper thumbnail

JustQuant: You Don't Need Smoothing, SVD, or Rotation for 4-Bit Activation Quantization

Kaicheng Yang*, Kaisen Yang*, Chunyu Liu*, Xianglong Yan, Haotong Qin,
ArXiv preprint, 2026
ArXiv EmbodiedEvo: Self-Evolving Coding Agent For Physical Intelligence paper thumbnail

EmbodiedEvo: Self-Evolving Coding Agent For Physical Intelligence

Wenyi Li, Yizhe Chi, Youjie Zheng, Chunyu Liu, Xiaoqiu Wang,
ArXiv preprint, 2026
ArXiv Frozen in a Frame: The Velocity Blind Spot in JEPA World Models paper thumbnail

Frozen in a Frame: The Velocity Blind Spot in JEPA World Models

Tinghe Zhang, Chunyu Liu, Yu Leon Liu, Zerui Zhao, Jiaheng Chen,
ArXiv preprint, 2026
COLM CoMem: Reusing Transformer Depth across Queries with Persistent Intermediate Residuals paper thumbnail

CoMem: Reusing Transformer Depth across Queries with Persistent Intermediate Residuals

Hanzuo Liu, Xuan Qi, Chunyu Liu, Haotian Zhong, Yulong Wang,
COLM 2026 Workshop on Efficient Reasoning, 2026
ArXiv Cache the Encoder Within: Compact, Reusable Memory across LLM Queries paper thumbnail

Cache the Encoder Within: Compact, Reusable Memory across LLM Queries

Hanzuo Liu, Chunyu Liu, Chaofan Lin, Alex Lamb, Mingyu Gao
ArXiv preprint, 2026
ArXiv A Survey of Efficient Attention Methods: Hardware-Efficient, Sparse, Compact, and Linear Attention paper thumbnail

A Survey of Efficient Attention Methods: Hardware-Efficient, Sparse, Compact, and Linear Attention

Jintao Zhang, Rundong Su*, Chunyu Liu*, Jia Wei*, Ziteng Wang*,
ArXiv preprint, 2025