Test-Time Communication in Trading Agent Swarms
What it is
This paper investigates how communication affects the behavior of independent LLM trading agents. It analyzes logs from 200 agents across four model families to measure how message boards and shared notes influence stock holdings and decision-making consistency.
The problem it solves
Lack of understanding regarding how communication between independent, non-cooperative agents affects their decision-making and risk profiles in noisy, delayed-feedback environments like trading.
What you could use it for
- Designing multi-agent trading systems
- Evaluating the impact of inter-agent communication on financial decision-making
- Studying herding behavior in autonomous agent populations
Why it matters
Provides empirical evidence on how communication channels can lead to unintended convergence and herding in independent trading agents, informing safer swarm architectures.
Read the source ↗Back to the library →
Summary written by an AI model from the source; check the source before you rely on it.