Skip to content

AI agent

Definition
AI-generated

An AI agent is a system that uses a language or multimodal model together with a control loop: it plans steps, calls tools (code, APIs, retrieval), observes results, and iterates until a goal is met—rather than producing a single static completion from one prompt.

Topics
LLM and Agents
Synonyms
LLM agent Autonomous agent (informal)
Plural
AI agents
Related
Human in the Loop (HITL) Harness Harness Engineering

Why it matters in GWAS

Agent-style setups can automate literature triage, pipeline wiring, or report generation, but they multiply failure modes (wrong tool args, silent retries); governance, logging, and human checkpoints matter as much as for any regulated analysis.

Example usage

"Our curation pipeline uses an AI agent that iteratively calls retrieval and validation tools to draft glossary entries, then submits uncertain cases for human review."

Human in the Loop (HITL) Harness Harness Engineering Genome-Wide Association Study (GWAS) Large Language Model (LLM) Model Context Protocol (MCP) Prompt Tool Calling
Related-term network

Click any node to open that term page.

Node order. The current term is first. Next come every page linked from Related terms, in the same top-to-bottom order as that list. Then new terms appear in two-hop layers: for each of those direct neighbors (in that same order), their related terms are appended if not already listed. A third pass does the same for two-hop nodes. The walk stops when a fixed maximum number of nodes is reached.

References

  • Yao S, et al. (2023). ReAct: Synergizing reasoning and acting in language models. arXiv:2210.03629.
← African Ancestry Akaike information criterion (AIC) →

Last updated (UTC · Git history)