← Search

Weilun Zhao

2 accepted papers

2026

DSSA: Dense-Sparse Switchable Attention for Seamless Short-to-Long Adaptation

ICLR 2026poster

Long-sequence processing is a critical capability for modern large language models. However, the self-attention mechanism in the standard Transformer architecture faces severe computational and memory bottlenecks when processing long sequences. While trainable sparse attention methods offer a promis…

Cited by 0SourceScholar
2025

FR-Spec: Accelerating Large-Vocabulary Language Models via Frequency-Ranked Speculative Sampling

ACL 2025long

Speculative sampling has emerged as an important technique for accelerating the auto-regressive generation process of large language models (LLMs) by utilizing a draft-then-verify mechanism to produce multiple tokens per forward pass. While state-of-the-art speculative sampling methods use only a si…