saved
SwarmWorld: Stigmergic technological evolution in societies of language-model agents
Hraness wrote this summary from a saved copy of the source. Quotations are taken word for word from the source.
gist
In SwarmWorld, from Subhadeep Pal, Fiona Y. Wang, and Markus J. Buehler, initially identical language-model agents build technology in a simulated world. They explore, process resources, build persistent artifacts, and write executable controllers, and a deterministic simulator decides what works, including after the agents are removed. In the preprint, shared societies form broader, more resilient technology portfolios than a strong isolated-search baseline, though isolated search stays competitive for the strongest artifact. Most reuse begins with agents observing the world rather than talking, and physical stigmergy, coordination through traces in the world, supports capable societies on its own.
ideas
- Split cognition from consequence. Agents propose architectures and controllers inside fixed schemas; the simulated world determines function, including under unseen disturbances after the agents are gone.
- Shared worlds win on portfolios, not always on the best object. Societies beat a strong best-of-N isolated-search baseline on breadth and resilience, while isolated search remains competitive for the strongest artifact.
- Physical stigmergy is enough. Most reuse begins by observing persistent artifacts rather than communicating; explicit culture amplifies collaboration, but functional benefits depend on outcome and timescale.
- Unassigned agents differentiate. Exploration, construction, maintenance, and coordination emerge without assigned roles or recipes, and those behaviors shift as the world matures.
- Interaction builds an ecology. Technologies accumulate through collaborative construction, executable inheritance, and agent-artifact networks rather than universally superior individual inventions.
quotes
“SwarmWorld splits cognition from consequence: agents propose architectures and controllers within fixed action and material schemas, while the simulated world determines function.”
“Shared societies develop broader, more resilient technological portfolios than a strong best-of-N isolated-search baseline”
“most reuse beginning through physical observation rather than communication”
“Physical stigmergy alone supports capable societies, while interaction drives persistent technological ecologies rather than universally superior individual inventions.”