Google shows neural cellular automata can do multi-step reasoning on decentralized hardware

tyrell_turing · x · 2026-10-24

• Google's Paradigms of Intelligence team (Blaise Aguera y Arcas et al., first author Mayalen Etcheverry) released the paper "Reasoning with Neural Cellular Automata (NCA)" on arXiv.
• Core question: can complex multi-step reasoning emerge purely from cells that only talk to their immediate neighbors?
• Answer: yes — spatial locality is not a barrier to complex multi-step reasoning. The work targets AI designed for decentralized hardware, where low-power communication and fault tolerance are critical requirements.
• The team shared the paper with a detailed explainer thread.

Original post →

More from Research

• MIT-IBM Computing Research Lab hiring summer intern for non-autoregressive architectures — alec_helbling · 2026-10-24
• 400 LoRA steps on Gemma 4 31B beat dedicated decision model Jev — vanstriendaniel · 2026-10-24
• ProgramBench tests models rebuilding ffmpeg from scratch; top model resolves just 18.5% — jyangballin · 2026-10-24
• NVIDIA Earth-2 simulates thousands of cyclone seasons, one in under 10 min — NVIDIA Developer · 2026-10-24
• BrickWright: agents build LEGO assemblies with perfect physical validity — cmu-gil · 2026-10-24
• A byte-level transformer works, but BPE exists for a reason: context length — max_paperclips · 2026-10-24

Research channel →