Zhiyuan Yu

TRACE: Trajectory Aware Reasoning for Multi-Turn Adversarial Conversation Evaluation

This work introduces Trace, a multi-turn defense with trajectory-aware structured reasoning that balances usability and safety, and trains Llama-3.1-8B-Instruct with SFT and GRPO under a multi-component reward that jointly optimizes helpfulness on benign prompts and robustness against jailbreak attempts.