activity
20242026
collaborators

7 papers

cs.LG2026

Faster Than Flash: Exploiting Attention Sparsity for Efficient Long-Context Decoding

Zhigeng Liu, Zhiyuan Ning, Ruixiao Li +5

The development of long-context Large Language Models (LLMs) is constrained by the memory bandwidth bottleneck and quadratic complexity of the attention mechanism during decoding.…

cs.CL2026

FourierSampler: Unlocking Non-Autoregressive Potential in Diffusion Language Models via Frequency-Guided Generation

Siyang He, Qiqi Wang, Xiaoran Liu +8

Despite the non-autoregressive potential of diffusion language models (dLLMs), existing decoding strategies demonstrate positional bias, failing to fully unlock the potential of ar…

cs.CL2025

Beyond Real: Imaginary Extension of Rotary Position Embeddings for Long-Context LLMs

Xiaoran Liu, Yuerong Song, Zhigeng Liu +6

Rotary Position Embeddings (RoPE) have become a standard for encoding sequence order in Large Language Models (LLMs) by applying rotations to query and key vectors in the complex p…

cs.CL2025

Sparse-dLLM: Accelerating Diffusion LLMs with Dynamic Cache Eviction

Yuerong Song, Xiaoran Liu, Ruixiao Li +5

Diffusion Large Language Models (dLLMs) enable breakthroughs in reasoning and parallel decoding but suffer from prohibitive quadratic computational complexity and memory overhead d…

cs.CL2025

Beyond Homogeneous Attention: Memory-Efficient LLMs via Fourier-Approximated KV Cache

Xiaoran Liu, Siyang He, Qiqi Wang +9

Large Language Models struggle with memory demands from the growing Key-Value (KV) cache as context lengths increase. Existing compression methods homogenize head dimensions or rel…

cs.CL2025

Thus Spake Long-Context Large Language Model

Xiaoran Liu, Ruixiao Li, Mianqiu Huang +11

Long context is an important topic in Natural Language Processing (NLP), running through the development of NLP architectures, and offers immense opportunities for Large Language M…