Preregistered 'physics from relations' simulation fails first large-scale run
tallmetommy · x · 2026-11-02
The author turned the hypothesis that "physical laws are the residue of mutually consistent relations" into a minimal model: no coordinates, distances, or pre-existing lattice—only relational graphs and a frozen microscopic energy rule—with preregistered success criteria (λ₂N^(-2/d) finite-size scaling, ds≈dH, dw→2) run at N=32/64/128/256 with 4 seeds.
Result: the low-energy regime recovers an exact cycle-like 1D phase, but no stable higher-dimensional continuum phase emerged, and parallel tempering completed zero full cold-to-hot round trips, failing the equilibrium-sampling requirement too. The honest verdict: the theory is neither confirmed nor dead—just not validated by the first large run.
More from Research
- HalachaBench sparks debate: is Talmudic Q&A testing knowledge or reasoning? — teortaxesTex · 2026-11-02
- Sebastian Raschka releases free 11-hour course on building reasoning models from scratch — AxSaucedo · 2026-11-02
- Microsoft and Berkeley open-source TokenEconomizer, cutting coding agent costs 18–31% — victor_explore · 2026-11-02
- TLA-Verus uses LLM agents to prove temporal properties of TLA+ specs — tianyin_xu · 2026-11-02
- J-Lens is just a first-order Taylor expansion, researcher points out — burny_tech · 2026-11-02
- Mollick: expert-level work benchmarks are brutally hard, and ARC-AGI, GDPval, METR are saturated — emollick · 2026-11-02