A Visual Guide to LLM Agents
Explore the main components of what makes LLM Agents special.
An agent is anything that can be viewed as perceiving its environment through sensors and acting upon that environment through actuators.
Agents interact with their environment and typically consist of several important components
chain-of-thought
This is where planning comes in. Planning in LLM Agents involves breaking a given task up into actionable steps.
reasoning” and “thinking” a bit loosely as we can argue whether this is human-like thinking or merely breaking the answer down to structured steps.
without any examples (zero-shot prompting)
Providing examples (also called few-shot prompting7)
ReAct
Reflecting
These Multi-Agent systems usually consist of specialized Agents, each equipped with their own toolset and overseen by a supervisor.
three LLM roles
SELF-REFINE
To enable planning in LLM Agents, let’s first look at the foundation of this technique, namely reasoning.
