الباحثون

Jinghao Pang

المنشورات 2

نسخة أولية وصول مفتوح

Can Agent Harnesses and Inference Engines Hear Each Other? The HEAR Protocol for Agentic LLM Serving

Jiaqi Zhao, Haodong Chen, Jitai Hao وآخرون · 2026

LLM agents increasingly execute complex workflows involving multi-turn reasoning, tool use, and parallel agents. Efficient serving requires decisions that span two layers with complementary information: the agent harness understands workflow dependencies, context lifecycles, and execution objectives, whereas the infere …

نسخة أولية وصول مفتوح

Beyond Refusal Patterns: Safe-Role Internalization for Robust and Generalizable LLM Safety Alignment

Jinghao Pang, Jitai Hao, Qiang Huang وآخرون · 2026

Large Language Models (LLMs) have achieved remarkable capabilities but remain vulnerable to jailbreak attacks that elicit harmful or unsafe outputs. Existing safety alignment approaches, including Supervised Fine-Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF), often require substantial attack-specif …

المؤلفون المشاركون