Overview
This project operationalizes the 2026 preprint "There is no self-evidence: A physics of emptiness realisation" as a source-anchored software artifact. Its central claim is that a finite agent can use a boundary for prediction while never obtaining evidence that the boundary is ontologically real, and the software separates three local artifact roles: formal sanity checks for source equations, positive-control-style finite mechanism checks, and discriminating tests that reject stronger readings when a control is perturbed. The formal layer maps the paper's quantum free-energy principle (qFEP) and quantum reference frame (QRF) equations into finite operational surrogates, bridging each paper equation to a specific software artifact. The computed artifacts are a suite of finite quantum-information and contextuality audits — spanning two-qubit separability and entanglement entropy, Bell and contextuality witnesses, thermodynamic and open-system dynamics, seeded quantum-trajectory sampling checked against exact solutions, and frame-covariance checks for the quantum reference frame relabelings — with explicit positive controls, negative controls, or boundary checks recorded where the corresponding artifact contract requires them. The software represents QRF deployments as policies over boundary-channel sectorisations, using the same finite bitstream under self/environment/contextual relabelings so that QRF labels can organize prediction, action selection, and transformation covariance while failing to become evidence for an ontological self/world boundary. Bayesian model reduction is implemented as a sweep over prior precision and metacognitive access, extended with a sensitivity grid over observation noise. The separation prior is pruned only when removing it lowers the model's free energy, and kept when its remaining contribution to accuracy still offsets its complexity cost. The active-inference layer uses the inferactively-pymdp library (Heins et al. 2022) with profile-specific likelihood, transition, preference, and prior arrays for the separation-constrained, opacified, and post-dual quantum reference frame deployments, then records posterior beliefs, policy posteriors, selected actions, expected-free-energy summaries, seeded stochastic ensembles with null controls and replay seeds, and confidence intervals, without treating those simulations as empirical subject data. Practice protocols, compassion-policy scope, criticality-style indicators, quantum-boundary dynamics, empirical adapters, and artifact-release readiness are therefore written as bounded model interfaces, simulated indicators, local private release-readiness records, or blocked evidence classes. A physical realization of the quantum free-energy principle, public independent reproduction, and any human practice efficacy, neural measurement, or clinical outcome remain blocked future evidence classes.
All manuscript and code source materials are available at https://github.com/docxology/realizing_emptiness
Overview source: Curated paper metadata.
Methods and contributions
Read the source for the full argument, qualifications, and evidence.
Findings and contributions
- Within the finite ledger, the same bitstream supports different QRF partitions, so the partition organizes the model without being evidence for its ontology.
- All admissible deployments preserve normalized boundary probability mass, while the perturbation control fails.
- In the surrogate model, the separation prior is useful only through agency.
- In the finite surrogate, Bayesian model reduction tracks the separation prior's net value and prunes it at high metacognitive access, at the weakest credible precision.
- The author explicitly does not claim that the software proves emptiness, removes a self, or realizes a physical quantum free-energy principle.
Methods
- Six-channel boundary screen read under three QRF sector relabelings — One b0–b5 bitstream is partitioned into separation-constrained, opacified and post-dual sector labelings to model QRF deployments.
- Equation registry mapping source qFEP/QRF equations to software artifacts — Equations from Sandved-Smith et al. (2026) are mapped into a fourteen-row registry and instantiated as finite QRF and BMR surrogates.
- Bayesian model reduction sweep over prior precision and metacognitive access — BMR is swept over separation-prior precision and metacognitive access, with an added sensitivity grid over observation noise.
- pymdp active-inference simulations with profile-specific A, B, C, D arrays — Uses inferactively-pymdp to run each QRF profile as an explicit generative model, recording posteriors, policies and seeded ensembles with null controls.
- Finite quantum-information and contextuality audits with controls — Runs finite checks such as two-qubit separability, Bell and contextuality witnesses, open-system dynamics and seeded trajectories checked against exact solutions.
Summary sources: Paper metadata and evidence · Extracted source text.
Document downloads
PDFs are hosted here; supporting sources may use pinned archives.
- realizing_emptiness_combined.pdf PDF · 14.98 MiB · extracted-text source
Citation
Citation metadata follows the unified bibliography.
Catalog details and resources
Related works
Ranked by recorded methods, topics, contributors and associated software, with contributor roles and domain-only matches labelled. Metadata overlap is not a citation or a scientific equivalence.