Picture for Zuchao Li

Zuchao Li

IAM: Efficient Inference through Attention Mapping between Different-scale LLMs

Add code
Jul 16, 2025
Viaarxiv icon

DAC: A Dynamic Attention-aware Approach for Task-Agnostic Prompt Compression

Add code
Jul 16, 2025
Viaarxiv icon

SpindleKV: A Novel KV Cache Reduction Method Balancing Both Shallow and Deep Layers

Add code
Jul 09, 2025
Viaarxiv icon

AMIA: Automatic Masking and Joint Intention Analysis Makes LVLMs Robust Jailbreak Defenders

Add code
May 30, 2025
Viaarxiv icon

Segment First or Comprehend First? Explore the Limit of Unsupervised Word Segmentation with Large Language Models

Add code
May 26, 2025
Viaarxiv icon

Faster MoE LLM Inference for Extremely Large Models

Add code
May 06, 2025
Viaarxiv icon

Model Hemorrhage and the Robustness Limits of Large Language Models

Add code
Mar 31, 2025
Viaarxiv icon

Label Drop for Multi-Aspect Relation Modeling in Universal Information Extraction

Add code
Feb 18, 2025
Viaarxiv icon

Joint Automatic Speech Recognition And Structure Learning For Better Speech Understanding

Add code
Jan 13, 2025
Figure 1 for Joint Automatic Speech Recognition And Structure Learning For Better Speech Understanding
Figure 2 for Joint Automatic Speech Recognition And Structure Learning For Better Speech Understanding
Figure 3 for Joint Automatic Speech Recognition And Structure Learning For Better Speech Understanding
Figure 4 for Joint Automatic Speech Recognition And Structure Learning For Better Speech Understanding
Viaarxiv icon

Imitate Before Detect: Aligning Machine Stylistic Preference for Machine-Revised Text Detection

Add code
Dec 22, 2024
Viaarxiv icon