Smartipedia

Attack Success Rate (AgentDojo)

Attack Success Rate in AgentDojo is a standardized metric measuring the percentage of successful adversarial attacks against AI agents, revealing that more capable models paradoxically face higher attack success rates while defensive measures can reduce vulnerability from 25% to 8%.

How do LLM Agents use an Imperative Loop?

LLM agents use imperative loops as their core architectural pattern, cycling through observe-think-act phases to break down complex tasks into manageable steps and work persistently toward goal completion using integrated tools and reasoning capabilities.

Imperative Loop (concise, simple english)

The imperative mood is a simple grammatical form that uses the base verb to give direct commands, instructions, or requests to someone.

Imperative Loop

Imperative loops are programming constructs that provide explicit control over repetitive execution through manual state management and sequential command specification, contrasting with declarative approaches that abstract away iteration mechanics.

State Machine Replication

State Machine Replication is a fundamental distributed computing technique that ensures fault tolerance by maintaining identical state across multiple server replicas through coordinated, deterministic operation processing.

Apache BookKeeper

Apache BookKeeper is an open-source distributed logging service that provides reliable, scalable, and low-latency storage for sequential data through a quorum-based replication system optimized for append-only workloads.

Tokenization

Tokenization is a versatile technique that replaces sensitive or complex data with non-sensitive placeholder values, serving critical roles in data security, natural language processing, and blockchain-based asset digitization.

dirty reads

Dirty reads are database concurrency anomalies where transactions read uncommitted data from other transactions, potentially leading to inconsistent results, though they can be deliberately allowed in specific scenarios where performance outweighs consistency requirements.

invariants

Invariants are properties or quantities that remain unchanged under specific transformations or operations, serving as fundamental tools across mathematics, computer science, and physics for understanding stability, proving correctness, and revealing underlying structural relationships.

ACID (computing)

ACID is a set of four properties (Atomicity, Consistency, Isolation, Durability) that ensure reliable and consistent database transaction processing, guaranteeing data integrity even in the presence of system failures or concurrent access.

AgentDojo

AgentDojo is an open-source evaluation framework developed by ETH Zurich and Invariant Labs that assesses the security and utility of AI agents, particularly their resilience to prompt injection attacks where malicious external data can hijack agent behavior.

Token Trajectories

Token trajectories represent the sequential paths that discrete computational units follow through AI systems, enabling more sophisticated tracking, analysis, and optimization of information flow in applications ranging from video understanding to autonomous driving.