15 papers
Breaking the Scale Barrier: One-Shot Knowledge Transfer via Frequency Transform
Jianlu Shen, Fu Feng, Yucheng Xie +2
Transferring knowledge by fine-tuning large-scale pre-trained networks has become a standard paradigm for downstream tasks, yet the knowledge of a pre-trained model is tightly coup…
SafeGene: Reusable Adapters for Transferable Safety Alignment
Yanghan Wang, Zhiqiang Kou, Fu Feng +2
Open-weight LLMs are increasingly fine-tuned into customized assistants, but downstream fine-tuning can weaken safety alignment and make models more vulnerable to malicious prompts…
Constraint-based Pre-training: From Structured Constraints to Scalable Model Initialization
Fu Feng, Yucheng Xie, Ruixiao Shi +2
The pre-training and fine-tuning paradigm has become the dominant approach for model adaptation. However, conventional pre-training typically yields models at a fixed scale, wherea…
SysOM-AI: Continuous Cross-Layer Performance Diagnosis for Production AI Training
Yusheng Zheng, Wenan Mao, Shuyi Cheng +8
Performance diagnosis in production-scale AI training is challenging because subtle OS-level issues can trigger cascading GPU delays and network slowdowns, degrading training effic…
A Creative Agent is Worth a 64-Token Template
Ruixiao Shi, Fu Feng, Yucheng Xie +3
Text-to-image (T2I) models have substantially improved image fidelity and prompt adherence, yet their creativity remains constrained by reliance on discrete natural language prompt…
A Unified Framework for Knowledge Transfer in Bidirectional Model Scaling
Jianlu Shen, Fu Feng, Jiaze Xu +3
Transferring pre-trained knowledge from a source model to a target model of a different architectural size is a key challenge for flexible and efficient model scaling. However, cur…