Picture for Xinchi Qiu

Xinchi Qiu

Don't Make It Up: Preserving Ignorance Awareness in LLM Fine-Tuning

Add code
Jun 17, 2025
Viaarxiv icon

DES-LOC: Desynced Low Communication Adaptive Optimizers for Training Foundation Models

Add code
May 28, 2025
Viaarxiv icon

Editing as Unlearning: Are Knowledge Editing Methods Strong Baselines for Large Language Model Unlearning?

Add code
May 26, 2025
Viaarxiv icon

LUNAR: LLM Unlearning via Neural Activation Redirection

Add code
Feb 11, 2025
Viaarxiv icon

Photon: Federated LLM Pre-Training

Add code
Nov 05, 2024
Figure 1 for Photon: Federated LLM Pre-Training
Figure 2 for Photon: Federated LLM Pre-Training
Figure 3 for Photon: Federated LLM Pre-Training
Figure 4 for Photon: Federated LLM Pre-Training
Viaarxiv icon

DEPT: Decoupled Embeddings for Pre-training Language Models

Add code
Oct 07, 2024
Figure 1 for DEPT: Decoupled Embeddings for Pre-training Language Models
Figure 2 for DEPT: Decoupled Embeddings for Pre-training Language Models
Figure 3 for DEPT: Decoupled Embeddings for Pre-training Language Models
Figure 4 for DEPT: Decoupled Embeddings for Pre-training Language Models
Viaarxiv icon

PISTOL: Dataset Compilation Pipeline for Structural Unlearning of LLMs

Add code
Jun 24, 2024
Figure 1 for PISTOL: Dataset Compilation Pipeline for Structural Unlearning of LLMs
Figure 2 for PISTOL: Dataset Compilation Pipeline for Structural Unlearning of LLMs
Figure 3 for PISTOL: Dataset Compilation Pipeline for Structural Unlearning of LLMs
Figure 4 for PISTOL: Dataset Compilation Pipeline for Structural Unlearning of LLMs
Viaarxiv icon

FLea: Addressing Data Scarcity and Label Skew in Federated Learning via Privacy-preserving Feature Augmentation

Add code
Jun 13, 2024
Viaarxiv icon

Sheaf HyperNetworks for Personalized Federated Learning

Add code
May 31, 2024
Figure 1 for Sheaf HyperNetworks for Personalized Federated Learning
Figure 2 for Sheaf HyperNetworks for Personalized Federated Learning
Figure 3 for Sheaf HyperNetworks for Personalized Federated Learning
Figure 4 for Sheaf HyperNetworks for Personalized Federated Learning
Viaarxiv icon

The Future of Large Language Model Pre-training is Federated

Add code
May 17, 2024
Figure 1 for The Future of Large Language Model Pre-training is Federated
Figure 2 for The Future of Large Language Model Pre-training is Federated
Figure 3 for The Future of Large Language Model Pre-training is Federated
Figure 4 for The Future of Large Language Model Pre-training is Federated
Viaarxiv icon