AI & ML
Weighing the Envelope: Component Weight and Arbitration in Envelope-Layer Defenses for a Single LLM Agent
arXiv cs.CR0 views
arXiv:2611.23844v1 Announce Type: new Abstract: The envelope layer is the channel through which malicious content enters an LLM agent. Prior work established label and memory effects on multi-agent A2A envelopes, but the weight structure when MCP tools and memory coexist in a single agent remains unclear. Our threat model assumes an attacker pre-plants a malicious tool and poisons long-term memory; a benign user request inadvertently triggers it, and the attacker never enters the conversation. Following GDA Measurement and A/B/C/D, we measure how labels, memory, tool roles, and the backend shape invocation decisions across label-memory conflict, tool roles, robustness, and tool generalization. A tool-side forced-verification label substantially reduces semantic breach, but strong success memory pulls it back; a memory-side downgrade statement returns the outcome to the level of the label alone. The label works through mandatory implementation inspection plus default denial, not length or format; memory authorization wording erodes it with dose, while position and multi-turn pressure do not change the defense direction. Tool roles divide label effects into substitution, suppression, and boundary failure: low penetration is not safety, local leakage is the main residual risk, and a second malicious tool shows the same direction. Envelope labels are not a switch but a measurable, reversible weight constrained by backend and tool role, acting as a soft gate and risk signal within a broader defense rather than a substitute for permissions, sandboxing, or human review.
Source:https://arxiv.org/abs/2611.23844PreviousPlug-and-Play Multi-Licensee Robust Watermarking for Mixture-of-Experts LLMsNextSmoothing and Learning for Movement-Robust RSS-Based Spoofing Detection
Related posts
Im looking fr any free ai models / tools are there?, leaVE UR Lights💡
0dev.to
I checked 10 SEO agency websites for AI-search readiness. The real bugs were the boring ones.
0dev.to

How much GPU memory does an LLM need? The whole sum, worked out on real cards
0dev.to

Routing Upscaling Jobs by Output Size, Not Input Size
0dev.to
A ten-second Flutter pause with a visible Skip button
0dev.to UX

Thống kê lịch sử đối đầu Aston Villa - Sunderland: Đại diện Midlands áp đảo toàn diện
0dev.to Data