ChatInject: Abusing Chat Templates for Prompt Injection in LLM Agents
Paper • 2509.22830 • Published • 5 •
2
HWANCHANG
HwanChang0106
AI & ML interests
None yet
Recent Activity
upvoted a paper about 3 hours ago
Learn from Weaknesses: Automated Domain Specialization for Small Computer-Use Agents upvoted a paper about 3 hours ago
Self-Improving Language Models with Bidirectional Evolutionary Search upvoted a paper about 3 hours ago
ResearchMath-14K: Scaling Research-Level Mathematics via AgentsOrganizations
None yet
commented a
paper 8 months ago