Research Logs

Detailed experiment logs behind the tutorials. These pages keep the claims, failed boundaries, formal checks, and implementation sketches separate from the beginner-facing exposition.

Scoped claims Replayable checks No hidden assumptions

Short paper: Fragment-sensitive qelim and safe table updates

A two-page PDF and compact academic-style note for readers who want the main results without the tutorial or the full research log: core equations, scoped claims, measured qelim result, safe pointwise revision, and exact limitations.

Energy-based optimizers and telemetry

Evidence behind Tutorial 59: ZenoEnergy verifier-call reduction, Tau route-telemetry results, ZenoDEX fast-router timings, and the advisory-ranking boundary that keeps verifiers authoritative.

Energy-based route telemetry for Tau

The research trail behind Tutorial 58: synthetic EBM feasibility, small-real-data negative evidence, expanded telemetry improvement, and the guard, abstention, validation, and fallback boundary.

Tau qelim and TABA table semantics

The neuro-symbolic research log behind the Tau optimization and TABA table work: fragment-sensitive qelim behind Tutorial 40, the measured TAU_QELIM_BACKEND=auto candidate, BDD existential abstraction, finite table proof artifacts, replayed Tau finite-table kernels, atomless cell projection, prefix-clopen carriers, safe infinite-recursive table evidence, feature-flagged Tau table helpers, the TauLang-Experiments demo workflow, protocol firewall demos, collateral reason routers, incident-memory tables, and the remaining frontier.

How to read these pages

A research log is not a tutorial and not a proof of unrestricted behavior. Each note states what was checked, what was only tested, what failed, and what remains outside the current model.