the
∞
.ai
Search
/
Concepts
trail ·
0 nodes
Every concept
Agents
8 concepts, 8 reviewed and merged.
Composition
Multi-Agent Systems
Rather than one model doing everything, give several a role each and let them talk: a writer and a critic, a planner…
verified
Control
ReAct
A model asked to reason produces a plausible chain and no way to check it.
verified
Self-Reflection
An agent that fails and retries the same way fails the same way.
verified
Task Decomposition
Some problems are too big to answer in one step and easy once split.
verified
Interface
Constrained Decoding
Asking a model for JSON and hoping is a coin flip at scale.
verified
Function Calling
A model that can only produce text cannot check the weather.
verified
Safety
Prompt Injection
A model reads instructions and data through the same channel, because to a language model there is no difference…
verified
State
Agent Memory
The context window is an agent's entire working memory, and it is small and expensive.
verified