Stanford Paper: More Information Can Hurt LLM Trading Agents

A Stanford team released an arXiv paper on empirical information design for multi-agent LLM systems, showing across experiments with 10 LLMs that giving some trading agents more information can sharply reduce their returns.

2026-11-08 ~ 2026-11-08 · 2 related posts

• Stanford paper: more info helps some LLM trading agents, wrecks others — rohanpaul_ai · 2026-11-08
• Stanford's multi-agent LLM information-design paper now on arXiv — rohanpaul_ai · 2026-11-08