Beijing

Junda Zhu

Advanced
@chuhac

LLM Trainer

Reasoning-to-Defend. [EMNLP 2025] Reasoning-to-Defend: Safety-Aware Reasoning Can Defend Large Language Models from Jailbreaking

12

ATM-RAG. [EMNLP 2024] ATM: Adversarial Tuning Multi-agent System Makes a Robust Retrieval-Augmented Generator

9

llm_hf_ds. use huggingface deepspeed api to sft large models with zero-1,2,3 and zero offload

1
3
Apply