CapLite: Prompt-Injection Defense for Small Language Model Agents without Model-Written Security Artifacts
cs.AI, cs.CL, cs.CR
Submitted: 2026-11-12
Updated: 2026-11-12
Terminology
Sources
- Design Patterns for Securing LLM Agents against Prompt Injections
- StruQ: Defending Against Prompt Injection with Structured Queries
- SecAlign: Defending Against Prompt Injection with Preference Optimization
- Meta-SecAlign: Training LLMs against Prompt Injection for Robust Agents
- Securing AI Agents with Information-Flow Control
- Defeating Prompt Injections by Design
- AgentDojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents
- The Granularity Mismatch in Agent Security: Argument-Level Provenance Solves Enforcement and Isolates the LLM Reasoning Bottleneck
- Gemma 4 Technical Report
- The Llama 3 Herd of Models
- Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection
- When Tool Outputs Become Commands: Separating Action Induction from Runtime Authorization in Tool-Augmented LLM Agents
- TinyLLM: Evaluation and Optimization of Small Language Models for Agentic Tasks on Edge Devices
- AttriGuard: Defeating Indirect Prompt Injection in LLM Agents via Causal Attribution of Tool Invocations
- Defending Against Indirect Prompt Injection Attacks With Spotlighting
- ACE: A Security Architecture for LLM-Integrated App Systems
- ToolFence: Fine-Grained Authorization for Secure Tool-Using LLM Agents
- Ministral 3
- MetaPermit: Scalable and Auditable Access Control for AI Agents via LLM-Inferred Meta-Attributes
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents