Machine Heart
Jul 8, 2026 · Information Security
Prompt Compression Creates a New LLM Vulnerability: HKUST’s Black‑Box Attack Framework COMA (ASE 2026)
The paper reveals that prompt‑compression modules, widely used to reduce token usage in LLM agents, can be exploited to erase critical safety constraints, enabling a black‑box attack called COMA that achieves up to 71% success across multiple compressors and tasks, and proposes isolated compression as an effective defense.
COMALLM securityadversarial attacks
0 likes · 10 min read
