Zirui Liu

9

Papers

9

Total Citations

Papers (9)

Zeroth-Order Fine-Tuning of LLMs with Transferable Static Sparsity

Longer Context, Deeper Thinking: Uncovering the Role of Long-Context Ability in Reasoning

NeurIPS 2025arXiv

Flexible Group Count Enables Hassle-Free Structured Pruning

TVE: Learning Meta-attribution for Transferable Vision Explainer

Soft Prompt Recovers Compressed LLMs, Transferably

GNNs Also Deserve Editing, and They Need It More Than Once

LLM Maybe LongLM: SelfExtend LLM Context Window Without Tuning

Knowledge Graphs Can be Learned with Just Intersection Features

KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache