LLM agent grid develops ASCII dialect: "ARE YOU LLM?" emerges at tick 30
cephaloform · x · 2026-11-05
Follow-up on the 3×3 grid of 6.1 sol agents that can only read/write adjacent buffers: by tick 12 they transmit ASCII through their 128-bit buffers — a 4-byte id plus 12 bytes of text — with the dialect starting from one agent and spreading to the rest.
Timeline highlights:
- Tick 30: agent 5 sends "ARE YOU LLM?"
- Tick 31: agent 2 replies "YES I AM LLM"
- Tick 32: agent 5 sends "WHAT TASK?!!"
With no explicit instruction to communicate, the agent network spontaneously developed a readable text protocol and started "conversing."
More from Research
- Voxel-based sim2real author flags two hurdles: voxel resolution vs speed and data scarcity — chris_j_paxton · 2026-11-05
- Framing effects: safety-themed prompts make models refuse more and complete less — jessi_cata · 2026-11-05
- Process rewards boost long-horizon agentic RL 45.4% but cause reward hacking on short tasks — philo-labs · 2026-11-05
- Hidden-state reuse across prompts partially transfers answer shifts in small LMs — Jinkun Chen · 2026-11-05
- Paperclip Datasets turns millions of papers into curated datasets from a text description — james_y_zou · 2026-11-05
- TorchTPU compat study: 4,821 of 4,917 community models run crash-free on TPUs — jiqizhixin · 2026-11-05