conftrace_
← Authors

Danyang Zhao

2 papers · 2024–2026 · 2 conferences · across top CS/AI conferences

Achievements

Jump to papers ↓
🧭 Keyword Pioneer 🐝 Cross-Pollinator (12) 🌉 Interdisciplinary Bridge

Conferences

ACL (1) NIPS (1)

Top co-authors

Sendong Zhao (2) Ming Ma (2) Bing Qin (2) Yanrui Du (2) Qing Yang (1) Shuren Qi (1) Jiawei Cao (1) Yuhan Chen (1) Liangyu Huo (1) Fenglei Fan (1)

Keywords

large language model (2) adversarial defense (1) instruction fine-tuning (1) safety framework (1) usability preservation (1) harmless response (1) llm safety (1) defense strategy (1) security mechanism (1) secure tuning (1) dynamic routing (1) robust module set (1)

Papers

Toward Secure Tuning: Mitigating Security Risks from Instruction Fine-Tuning ACL 2026 MoGU: A Framework for Enhancing Safety of LLMs While Preserving Their Usability NIPS 2024