TS-PEFT: Unveiling Token-Level Redundancy in Parameter-Efficient Fine-Tuning
A token-level sparsity perspective for reducing redundant updates in parameter-efficient fine-tuning.
PEFT Token Sparsity LLM Fine-Tuning
Publications
You can also find my articles on my Google Scholar profile.
2026
NSR-Boost: A Neuro-Symbolic Residual Boosting Framework for Industrial Legacy Models
A neuro-symbolic residual boosting framework for upgrading industrial legacy models without full retraining.
Neuro-Symbolic AI Residual Boosting Legacy Models
Single pass Poisson disk sampling via circle packing
A single-pass Poisson-disk sampling method via circle packing for efficient blue-noise sample generation.
Poisson-Disk Sampling Circle Packing Blue Noise
Stratos: An End-to-End Distillation Pipeline for Customized LLMs under Distributed Cloud Environments
An end-to-end distillation pipeline for customized LLM deployment under distributed cloud constraints.
LLM Distillation Distributed Cloud Model Deployment
2025
ACME: Adaptive Customization of Large Models via Distributed Systems
A distributed customization system for adapting large models while reducing centralized training cost.
LLM Customization Distributed Systems Personalized Models
2024
Multi-Granularity Federated Learning by Graph-Partitioning
A graph-partitioning multi-granularity federated learning method for efficient and robust edge collaboration.
Federated Learning Graph Partitioning Edge Intelligence
2023
MG²FL: Multi-Granularity Grouping-Based Federated Learning in Green Edge Computing Systems
A multi-granularity grouping framework for improving federated learning efficiency in green edge systems.
Federated Learning Green Edge Computing Multi-Granularity
AImers-6G: AI-Driven Region-Temporal Resource Provisioning for 6G Immersive Services
An AI-driven resource provisioning approach for 6G immersive services across region-temporal demand.
6G Services Resource Provisioning Immersive Computing