Manuscript index
Papers & supporting notes
47 papers and notes in 16 research groups: 8 research manuscripts, 8 technical reports and 31 research notes. 47 have reading copies to read online or as PDF. Entries also link 9 narrated video overviews, 16 code, verification or evidence downloads and exact LaTeX source for 41 entries (under Technical material). These are working manuscripts, reports and notes, not journal publications or claims of external peer review.
Choose a reading format: Read Online provides rendered text and mathematics. Read PDF opens the page-faithful PDF reader; Download PDF saves the file. About explains the paper. Main papers appear first in each group; expand related notes for addenda and technical material.
Local research release: edition status, full readers and broader supporting reports →
Versions, formats, and accessibility
About these versions: Reading-copy numbers identify fixed PDF versions selected for this site. The October 3 copies corrected author attribution or metadata; the October 5 copies add the minor source clarifications and copyedits from the internal audit. The stated mathematical results are unchanged. Each entry also names its source manuscript revision. For questions or an unavailable copy, email Brian@B-Theory.com.
Reading formats: Online manuscripts use HTML + MathML with headings, equations and references. The PDF reader preserves the fixed page layout; the download provides the same PDF file. Edition notices and supporting records identify corrections, later evidence and scope limitations. Exact sources and videos remain available as additional links. Screen-reader behavior needs specialist testing; original PDFs are not tagged for assistive technology. Extracted-text companions have equation limitations.
Current manuscript lines
ManuscriptCorrespondence and Logical Matrices: A Typed Boolean Operator Calculus
Contribution: Defines a typed symbolic and numeric Boolean operator calculus and states compatibility identities connecting the two representations.
Technical material
Research status, version, and suggested sections
Where to look: See “Formula-valued logical matrices” and “Compatibility identities” in the September 2026 revision.
Research status: Foundations and synthesis manuscript; current revised version, September 2026.
Source revision: September 27 revised manuscript; earlier September 22 and September 27 revisions retained in the collection.
Site release: Reading copy 1 · PDF byline: Brian Theory
Related notes, addenda and supporting material (10)
Supporting methods noteGuarded Boolean Operators: Exact Dispatch and a Polynomial Guard-Degree Tradeoff
Contribution: A complete four-operator example shows how the permitted tests, their polynomial degree, the chosen coordinates and zero handling change exact operator identification.
Technical material
Reading scope: Nonempty certified guards, a promised zero-free domain and an explicit coordinate/observation language. No runtime speedup, learned-model result or cognitive claim.
Research status, version, and suggested sections
Where to look: Sections 2–3 give the four-operator model and degree/test tradeoff; Sections 4–5 separate zero-aware certification, test selection and faithful CM/LM representations.
Research status: September 28 reviewed methods/expository note. Review passes were conducted by the same assistant; external human review and priority remain open.
Source revision: September 28 reviewed source; exact 14-page PDF and TeX preserved. October 13 extended working draft is a pending successor, not part of this reading copy.
Site release: Reading copy 1 · PDF byline: Brian Theory
Supporting correction reportCorrectness Audit of Boolean CM Modal Models
Contribution: Distinguishes native Boolean logic, rotation operators, shared modules and independent modal registers; records corrected resource, readout and probability claims.
Technical material
Reading scope: Supporting correction provenance, not a replacement for the current foundations paper or blanket certification of later modal results. Historical timing-manifest discrepancy retained.
Research status, version, and suggested sections
Where to look: Start with the new correction ledger; original Sections 3–6 separate the models, Sections 9–14 qualify protocols and readout, and final sections state limits.
Research status: Historical September 18 audit available as website supporting material; internal AI/computational evidence, not external peer review.
Source revision: Verified baseline 1.0, 18 September 2026; unchanged 28-page PDF and exact modular/standalone source. New summary and current-relevance ledger dated 12 November 2026.
Site release: Reading copy 1 · PDF byline: Prepared for Brian / Quantum CM project; original byline preserved.
Methods and negative-results reportOperator-Level Boolean Computation with Correspondence Matrices
Contribution: Operator-level normalization and executable computation with explicit baseline and cost contracts.
Technical material
Reading scope: Current October 13 maintained source; completed studies retain failed utility gates and direct-packed wins. No generic runtime advantage is established.
Research status, version, and suggested sections
Where to look: Start with the edition notice and supporting records, then read the complete original pages or exact source.
Research status: Current October 13 maintained source; completed studies retain failed utility gates and direct-packed wins. No generic runtime advantage is established. Internal historical review records are distinct from external peer review.
Source revision: Current October 13 maintained source; completed studies retain failed utility gates and direct-packed wins. No generic runtime advantage is established.
Site release: Reading copy 1 · PDF byline: Original byline and date preserved as printed in the manuscript/source.
Specialist working manuscriptContextuality and Exact Information Resources over Finite Chain Rings
Contribution: Contextuality and exact information resources in a declared finite-chain-ring modal model.
Technical material
Reading scope: September 26 revision. Primitive bases, allowed transformations and possible/impossible readout are part of the model, not physical quantum assumptions.
Research status, version, and suggested sections
Where to look: Start with the edition notice and supporting records, then read the complete original pages or exact source.
Research status: September 26 revision. Primitive bases, allowed transformations and possible/impossible readout are part of the model, not physical quantum assumptions. Internal historical review records are distinct from external peer review.
Source revision: September 26 revision. Primitive bases, allowed transformations and possible/impossible readout are part of the model, not physical quantum assumptions.
Site release: Reading copy 1 · PDF byline: Original byline and date preserved as printed in the manuscript/source.
Specialist research noteExact Bernstein–Vazirani Query Complexity in Characteristic-Two Modal Computation
Contribution: Exact Bernstein–Vazirani query bounds under an explicit characteristic-two modal oracle/readout contract.
Technical material
Reading scope: September 27 reviewed source. Adaptive complete recorded readout and the pointwise oracle contract are essential; no quantum-device or cognitive advantage is inferred.
Research status, version, and suggested sections
Where to look: Start with the edition notice and supporting records, then read the complete original pages or exact source.
Research status: September 27 reviewed source. Adaptive complete recorded readout and the pointwise oracle contract are essential; no quantum-device or cognitive advantage is inferred. Internal historical review records are distinct from external peer review.
Source revision: September 27 reviewed source. Adaptive complete recorded readout and the pointwise oracle contract are essential; no quantum-device or cognitive advantage is inferred.
Site release: Reading copy 1 · PDF byline: Original byline and date preserved as printed in the manuscript/source.
Supporting technical companionSigned and Integral Lifts of Correspondence Matrices
Contribution: Scalar-change boundaries connecting Boolean phase labels to signed and integral representations.
Technical material
Reading scope: September 27 submission revision. The successful signed construction changes scalars; it is not a native Boolean complex amplitude theory or speedup.
Research status, version, and suggested sections
Where to look: Start with the edition notice and supporting records, then read the complete original pages or exact source.
Research status: September 27 submission revision. The successful signed construction changes scalars; it is not a native Boolean complex amplitude theory or speedup. Internal historical review records are distinct from external peer review.
Source revision: September 27 submission revision. The successful signed construction changes scalars; it is not a native Boolean complex amplitude theory or speedup.
Site release: Reading copy 1 · PDF byline: Original byline and date preserved as printed in the manuscript/source.
Supporting technical companionBoolean CM Phase Algebra: Native Operations, Tags, and Readout
Contribution: Native binary realizations, nilpotent tags and exact detector bounds under declared operation/readout assumptions.
Technical material
Reading scope: September 28 rewrite. Literal coefficient inspection and paid modal query/readout contracts must not be conflated.
Research status, version, and suggested sections
Where to look: Start with the edition notice and supporting records, then read the complete original pages or exact source.
Research status: September 28 rewrite. Literal coefficient inspection and paid modal query/readout contracts must not be conflated. Internal historical review records are distinct from external peer review.
Source revision: September 28 rewrite. Literal coefficient inspection and paid modal query/readout contracts must not be conflated.
Site release: Reading copy 1 · PDF byline: Original byline and date preserved as printed in the manuscript/source.
Supporting process companionProcess Closure and Restricted Conversion in a Chain-Ring Modal Model
Contribution: Composition, selective conditioning and restricted conversion boundaries for a chain-ring modal model.
Technical material
Reading scope: September 20 companion. Static support is not automatically closed process semantics. Read alongside Chain-Ring Contextuality and the correctness audit.
Research status, version, and suggested sections
Where to look: Start with the edition notice and supporting records, then read the complete original pages or exact source.
Research status: September 20 companion. Static support is not automatically closed process semantics. Read alongside Chain-Ring Contextuality and the correctness audit. Internal historical review records are distinct from external peer review.
Source revision: September 20 companion. Static support is not automatically closed process semantics. Read alongside Chain-Ring Contextuality and the correctness audit.
Site release: Reading copy 1 · PDF byline: Original byline and date preserved as printed in the manuscript/source.
Supporting bridge reportDoes the Symbolic LM Calculus Canonically Induce the Modal Measurement Theory?
Contribution: Separates Boolean valuation from the scalar enrichment and operational assumptions required by modal measurement.
Technical material
Reading scope: Recovered bridge/synthesis report dated 18 September 2026. Checked typeset reading copy with documented typography-only changes; exact original source retained. No new priority claim is made.
Research status, version, and suggested sections
Where to look: Start with the edition notice and supporting records, then read the complete original pages or exact source.
Research status: Recovered bridge/synthesis report dated 18 September 2026. Checked typeset reading copy with documented typography-only changes; exact original source retained. No new priority claim is made. Internal historical review records are distinct from external peer review.
Source revision: Recovered bridge/synthesis report dated 18 September 2026. Checked typeset reading copy with documented typography-only changes; exact original source retained. No new priority claim is made.
Site release: Reading copy 1 · PDF byline: Original byline and date preserved as printed in the manuscript/source.
Working report with supplementsExact Boolean Decomposition Artifacts under Conditioning: Algebraic Contracts and a Reproducibility Audit
Contribution: Conditioning, persistent representations and decomposition under explicit algebraic and storage contracts.
Technical material
Reading scope: Draft 3 is preserved alongside later convergence evidence. The later work is not silently merged into a final manuscript; storage formulas and correctness do not establish best-baseline speedups.
Research status, version, and suggested sections
Where to look: Start with the edition notice and supporting records, then read the complete original pages or exact source.
Research status: Draft 3 is preserved alongside later convergence evidence. The later work is not silently merged into a final manuscript; storage formulas and correctness do not establish best-baseline speedups. Internal historical review records are distinct from external peer review.
Source revision: Draft 3 is preserved alongside later convergence evidence. The later work is not silently merged into a final manuscript; storage formulas and correctness do not establish best-baseline speedups.
Site release: Reading copy 1 · PDF byline: Original byline and date preserved as printed in the manuscript/source.
Expository manuscriptObservation Topologies for Finite Propositional Semantics: Distinguishability, Distance, and Inference
Contribution: Shows how chosen propositional observations induce truth-signature partitions, finite observation spaces, and tests of distinguishability.
Technical material
Research status, version, and suggested sections
Where to look: See “Valuations, truth regions, and partitions” and “Observation-generated spaces” in v0.9.
Research status: Expository synthesis of standard finite constructions; current v0.9 manuscript.
Source revision: v0.9; sentence-fragment and truth-pattern example copyedits October 5, 2026. Earlier v0.6, v0.8, and v0.8.1 records retained in the collection.
Site release: Reading copy 2 · PDF byline: Brian Theory
Research manuscriptBinary Order Thinning of Finite Posets: Homology, Collapses, and Sharp Limits
Contribution: Gives a rooted-forest criterion for vanishing relative homology and collapse under one Boolean bit in height-at-most-two finite orders.
Technical material
Research status, version, and suggested sections
Where to look: See “The exact forest/collapse theorem” and the two “Sharpness” sections in local v3.
Research status: Focused research manuscript; internally reviewed local v3. Vertex minimality through six states is computer-assisted.
Source revision: Reviewed local v3, September 27, 2026; editorial domain, program-description, and supplement-access polish October 5, 2026. Stated results unchanged.
Site release: Reading copy 3 · PDF byline: Brian Theory
Related notes, addenda and supporting material (1)
Technical research addendumSigned incidence and a nine-state torsion witness in simultaneous two-bit order thinning
Contribution: Gives an exact integral and characteristic-two collapse criterion in a restricted two-bit regime, and realizes relative two-torsion with nine states even though both endpoint complexes have circle homotopy type.
Technical material
Reading scope: Finite posets of height at most two, with no triangle losing all three edges. Nine is an upper bound, not a minimum. Structural admissibility is not physical safety or predictive performance.
Research status, version, and suggested sections
Where to look: Sections 2-3 define the signed-incidence restriction and prove recognition; Sections 4-5 give the nine-state witness and inclusion map; Section 6 separates structure from prediction, cost and execution.
Research status: Two new independent internal AI proof and adversarial reviewers read and hashed the complete final source on 9 November 2026; no supported fatal, major or required minor mathematical correction remained. Human specialist review and historical priority remain open.
Source revision: Technical addendum v1.0 working release candidate, 9 November 2026; exact release source and seven-page PDF preserved. The parent manuscript remains unchanged.
Site release: Reading copy 1 · PDF byline: Brian Theory
Research draftSuccessive Concept Differentiation as Observation Refinement
Contribution: Separates signature refinement, carrier splitting, and order thinning, then studies the topology and homology of finite refinement constructions.
Technical material
Research status, version, and suggested sections
Where to look: See “Observation systems and three kinds of refinement” and “Carrier splitting under a new Boolean observation” in v0.5.
Research status: Integrated MIND research draft v0.5, with internal referee repairs; priority and external review remain open.
Source revision: Referee-repair draft v0.5, September 29, 2026; author corrected October 3, 2026; layer, nerve, and skeleton-proof polish October 5, 2026. Earlier revisions retained in the collection.
Site release: Reading copy 3 · PDF byline: Brian Theory
Related notes, addenda and supporting material (2)
Technical research addendumIntegral secondary maps in an observation four-cycle
Contribution: Realizes every integer multiplier of magnitude at least two in a single exclusive four-cycle with contractible coarse complex and torsion-free mixed-core link homology, and proves a zero-or-primitive obstruction in source row zero.
Technical material
Reading scope: Actual Boolean-observation refinements with exclusive graph exactly one K2,2; the row-zero theorem permits isolated exclusive vertices and nontrivial first-page quotients. No minimum-state, larger-graph, refined homotopy-type or cognitive claim.
Research status, version, and suggested sections
Where to look: Section 2 defines the observation product and filtration; Sections 3-4 prove the construction and integral index; Section 5 gives the separate 35-state example; Section 6 proves the row-zero boundary.
Research status: Version 1.0, 9 November 2026. Two independent internal AI proof and hostile reviews closed minor revisions with no supported major objection. Not external human peer review; specialist review and historical priority remain open.
Source revision: Technical addendum v1.0, 9 November 2026; exact reviewed source and seven-page PDF preserved. The fixed Observation Refinement referee-repair v0.5 dependency and its public manuscript remain unchanged.
Site release: Reading copy 1 · PDF byline: Brian Theory
Corrected research brief and historical notesObservation Refinement Beyond a Fixed Carrier, with Constrained Observation Design
Contribution: Fiber-profile and observation-design research with explicit correction history and surviving open questions.
Technical material
Reading scope: Unfinished direction. The old equality odim_P(Q)=dim_2(Q) is false and its dependent complexity argument is unsupported. The zero-byte v0.3 source is not reconstructed. Historical notes are not the current integrated MIND v0.5.
Research status, version, and suggested sections
Where to look: Start with the edition notice and supporting records, then read the complete original pages or exact source.
Research status: Unfinished direction. The old equality odim_P(Q)=dim_2(Q) is false and its dependent complexity argument is unsupported. The zero-byte v0.3 source is not reconstructed. Historical notes are not the current integrated MIND v0.5. Internal historical review records are distinct from external peer review.
Source revision: Unfinished direction. The old equality odim_P(Q)=dim_2(Q) is false and its dependent complexity argument is unsupported. The zero-byte v0.3 source is not reconstructed. Historical notes are not the current integrated MIND v0.5.
Site release: Reading copy 1 · PDF byline: Original byline and date preserved as printed in the manuscript/source.
Research noteHierarchical Concepts Under Finite Resolution: Observation Quotients and Adaptive Symmetry
Contribution: Defines observation quotients and resolution towers for ultrametric hierarchies while distinguishing symmetry measures from information gain.
Technical material
Research status, version, and suggested sections
Where to look: See “A coordinate-free observation quotient” and “Information gain is not a symmetry index” in v0.3.
Research status: Short synthesis and correction note; current v0.3 research copy.
Source revision: Internal fresh-referee v0.3, September 29, 2026; author corrected October 3, 2026. Original PDF retained in the local history.
Site release: Reading copy 2 · PDF byline: Brian Theory
Related notes, addenda and supporting material (1)
Methods companionWhen to Split and When to Merge Task-driven observation concepts with paid audits and repeated revision
Contribution: Finite-class risk certificates, paid-query stopping and repeated-revision examples, retaining small, zero and harmful policy gains.
Technical material
Reading scope: A finite representation library, observable histories and declared supervised validation. Transfer needs a quantitative distribution-change bound.
Research status, version, and suggested sections
Where to look: Read the observable interface and task-dependent distinction example, then the certificate cost study. For repeated revision, compare the rational policy envelopes with every sensitivity and frozen-policy loss case, not just the favorable reference setting.
Research status: AI-assisted research draft; internal AI checks documented. External human review and novelty assessment remain pending.
Source revision: 13 November 2026 manuscript with localized corrections and approved author credit; exact edition hashes linked below.
Site release: Reading copy 1 · PDF byline: Brian Theory · AI-assisted research draft
Research collection and working noteStructured Boolean Factorization
Contribution: Structural factorization mathematics with corrected Phase 5B proofs and retained negative computational evidence.
Technical material
Reading scope: The original Phase 5 note is historical within this collection. Read the later corrected proofs and negative-result gate first; no integrated final manuscript or algorithmic speedup is asserted.
Research status, version, and suggested sections
Where to look: Start with the edition notice and supporting records, then read the complete original pages or exact source.
Research status: The original Phase 5 note is historical within this collection. Read the later corrected proofs and negative-result gate first; no integrated final manuscript or algorithmic speedup is asserted. Internal historical review records are distinct from external peer review.
Source revision: The original Phase 5 note is historical within this collection. Read the later corrected proofs and negative-result gate first; no integrated final manuscript or algorithmic speedup is asserted.
Site release: Reading copy 1 · PDF byline: Original byline and date preserved as printed in the manuscript/source.
Technical reportFrom Observable States to Acquirable Predictive Models
Contribution: Characterizes reportable targets in a restricted finite ordered model, compares attainable distinctions with predictive equivalence, and treats alternating queries and known actions under an explicit finite-horizon contract.
Technical material
Research status, version, and suggested sections
Where to look: Start with Section 1; see Theorem 2 and Corollaries 3 and 5 for fixed-state acquisition, Example 8 for query orientation, and Section 7 for actions and timing.
Research status: Consolidated technical report, revised October 7, 2026. Internal AI audits are not external peer review; priority of the specialized classification remains unresolved.
Source revision: Revised technical report, October 7, 2026; site PDF adds metadata and reference-line wrapping without changing mathematical content.
Site release: Reading copy 1 · PDF byline: Brian Theory
Related notes, addenda and supporting material (6)
Technical addendumWhen Delayed Threshold Sensing Cannot Help: An Exact Scalar Acquisition-Timing Addendum
Contribution: Classifies the exact worst-case probe optimum for a disturbed scalar system with the original three-input alphabet, and proves a one-added-input feasibility separation without changing the query or deadline contract.
Reading scope: Fixed controls; exact continuous state; adversarial bounded disturbance; noiseless instantaneous equal-cost threshold queries; mandatory next-step zero-error forecasts; hard worst-case probe budgets. No finite-to-finite improvement, general active-diagnosis theory, empirical validation or consciousness conclusion is claimed.
Research status, version, and suggested sections
Where to look: Section 2 defines the operational contract; Section 3 gives the information bound; Theorem 4 gives complete equality; Theorem 5 gives the one-input extension; Section 6 records reproduction and limits.
Research status: Technical addendum v1.1, 9 November 2026. Internal AI proof and hostile reviews closed their recorded objections. Not human peer review; historical priority remains unresolved.
Source revision: Seven-page technical addendum v1.1, 9 November 2026. Exact reviewed LaTeX, PDF and supplied public archive preserved byte-for-byte. The online reader adds navigation, source TeX annotations and recorded MathML presentation repairs.
Site release: Reading copy 1 · PDF byline: Prepared for Brian Theory / AI-assisted mathematical development and internal review. Original manuscript byline retained.
Supporting technical addendumSequential Certificate Risk for Declared RF Carriers: A finite-horizon technical addendum
Contribution: Gives first-exclusion accounting, a complete finite-history conditional-risk linear program, and exact small-model comparisons between specified carrier-admissibility contracts under known fresh binary-symmetric noise.
Reading scope: Complete supporting technical addendum, including proofs, finite frontier appendix and references. Fixed state, known positive independent binary-symmetric noise, declared carriers and explicit public-history conditioning. No general all-horizon theorem, physical-safety claim or consciousness conclusion.
Research status, version, and suggested sections
Where to look: Sections 2-4 specify the model and conditional-risk contract; Sections 5-7 give exact benchmarks and gate comparisons; Section 8 states corrections and limits; Appendix A retains the supporting finite frontiers.
Research status: Technical addendum v0.1, 9 November 2026. Independent internal AI proof and hostile reviews closed their requested repairs against the final source. Not human peer review, proof-assistant certification or priority clearance. Authorship approval and physical applicability remain unresolved in the frozen source.
Source revision: Reviewed release candidate 0.1, 9 November 2026. The eight-page PDF and LaTeX are unchanged; the site reader adds navigation, MathML and source-formula fallbacks. Public metadata received whitespace-only corrections with refreshed hashes.
Site release: Reading copy 1 · PDF byline: Prepared for Brian Theory; AI-assisted derivation and internal mathematical review.
Explanatory appendixShared Clocks and Historical Targets
Contribution: Synthesizes existing MIND and acquisition results with a shared-clock constraint model and a fully specified eight-history archive-access example. No standalone new theory or general planning method is claimed.
Reading scope: Finite declared trajectories; one offset per module; explicit clock lifetimes and body width; eight restricted histories; fixed serialized request/response frames; preexisting archives and trusted reference; matched input ownership and controller rights. No consciousness, architectural superiority or unrestricted scalability claim.
Research status, version, and suggested sections
Where to look: Section A gives trajectory compatibility, shared-offset constraints and a qualified elimination bound. Section B gives exact archive-query costs and distinguishes access limits from complete-state erasure. Section C records verification and scope.
Research status: Explanatory appendix v0.2, 9 November 2026. Separate internal AI mathematical and hostile reviews passed against the exact source. Not external peer review; priority is not established.
Source revision: Five-page explanatory appendix v0.2, 9 November 2026. Exact source, PDF and approved public ZIP preserved byte-for-byte. The reader supplies navigation, source TeX annotations and recorded presentation repairs.
Site release: Reading copy 1 · PDF byline: B-Theory research programme
Technical noteAcquisition, Forgetting, and Target Change: Information Accounting for Adaptive Systems
Contribution: An exact acquisition/forgetting identity and rational finite-memory tracking benchmarks distinguish sensing, compression and target change.
Technical material
Reading scope: Finite targets, declared pre-observation records, priced sensing and persistent finite memory. The periodic controller is an achievable upper bound, not a global optimum.
Research status, version, and suggested sections
Where to look: Start with the accounting identity and worked counterexamples, then read the tracking and uniform-comparison extensions. Compare the finite synthesis recurrence with the explicit sensing and memory prices before interpreting the plots.
Research status: AI-assisted research draft; internal AI checks documented. External human review and novelty assessment remain pending.
Source revision: 13 November 2026 manuscript with localized corrections and approved author credit; exact edition hashes linked below.
Site release: Reading copy 1 · PDF byline: Brian Theory · AI-assisted research draft
Technical noteAcquiring a Predictive Capability through Restricted Noisy Experiments Exact frontiers and certified asymmetric irreversible acquisition
Contribution: Exact finite minimax prediction frontiers and rational certificates for restricted, asymmetric and irreversible experiment protocols.
Technical material
Reading scope: Finite known model families and supplied likelihoods/permissions. Prediction is distinct from historical parameter identification; general stochastic design and novelty remain open.
Research status, version, and suggested sections
Where to look: Read the transcript obstruction and exactly solved restricted experiment first, including Theorem 4.2 and its tie convention. Continue to the asymmetric extension and inspect both upper-policy and lower-prior certificates.
Research status: AI-assisted research draft; internal AI checks documented. External human review and novelty assessment remain pending.
Source revision: 13 November 2026 manuscript with localized corrections and approved author credit; exact edition hashes linked below.
Site release: Reading copy 1 · PDF byline: Brian Theory · AI-assisted research draft
Technical noteMemory–Prediction Tradeoffs for a Shared Binary Mechanism: Noisy Random Access, Adaptive Sampling, and Executable State Costs
Contribution: Exact noisy random-access and adaptive-sampling examples with explicit terminal, streaming and controller-code costs.
Technical material
Reading scope: Independent fair mechanism bits, stated channel noise and late coordinate queries. A 19-tag controller is a witness, not a proven minimum.
Research status, version, and suggested sections
Where to look: Read the quantization reduction before the adaptive separations. Check the corrected port numbering against each allocation and read executable state costs separately from terminal information bits. Keep the open minimum-state question distinct from a successful implementation witness.
Research status: AI-assisted research draft; internal AI checks documented. External human review and novelty assessment remain pending.
Source revision: 13 November 2026 manuscript with localized corrections and approved author credit; exact edition hashes linked below.
Site release: Reading copy 1 · PDF byline: Brian Theory · AI-assisted research draft
Technical reportFrom Observable Answers to Executable Queries for CM/LM Programs
Contribution: Connects finite information sufficiency, costed static selection and legal adaptive acquisition, with a pointwise addressing bound and an independent policy verifier.
Technical material
Research status, version, and suggested sections
Where to look: Sections 2–5 for model and proofs; Section 6 for CM/LM; Sections 7–9 for execution, verification and reproducibility.
Research status: Technical report v1.1, internally reviewed with AI assistance; human peer review and comprehensive novelty assessment remain open.
Source revision: Reviewed seven-page technical report v1.1, 8 November 2026. Original PDF and source preserved byte-for-byte.
Site release: Reading copy 1 · PDF byline: Brian Theory
Related notes, addenda and supporting material (1)
Completed bounded resultThe MIND: Five Finite Constructions — P2 mathematical addendum
Contribution: Gives the exact attained mutual-information interval for full binary support with fixed marginals and a positive cell floor; characterizes proper supports, aligned rational grids, intervention counterexamples and admissible local recodings.
Technical material
Reading scope: Complete 15-page combined MIND manuscript, including M1–M5, P1, P2, all proofs, tables and references. This entry highlights P2 in Section 9; the original P2-only reader remains available. The unchanged source and PDF retain their original edition and byline.
Research status, version, and suggested sections
Where to look: Read the full combined MIND manuscript; use its jump link to start at Section 9 for P2.
Research status: Result completed October 5, 2026. Elementary specialization of established coupling mathematics; analytic proofs and bounded internal AI-assisted checks, not external peer review or a novelty claim.
Source revision: First research edition with P1 and P2 mathematical addenda, October 5, 2026; site section reading copy dated October 10, 2026.
Site release: Reading copy 1 · PDF byline: Research prepared for Brian Theory; AI-assisted mathematical synthesis; not external peer review. PDF Author metadata is blank and preserved.
Finite-state pilotWhen Does an Explicit Self-Model Help? Calibration and Control under Internal Change in a Finite-State Pilot
Contribution: Separates Bayesian evidence use from architectural labeling; the explicit, compiled, and external trackers share the same mapping and outperform two specified restricted rules in the supplied benchmark.
Technical material
Reading scope: Full manuscript. Results are inherited source evidence; the reading copy does not independently rerun the benchmark or establish subjective awareness.
Research status, version, and suggested sections
Where to look: Start with Section 1 for the two questions, Section 3 for timing and the worked Bayesian update, Section 4 for the matched trio, and Sections 6 and 9 for results and limitations.
Research status: Analytical manuscript based on an inherited finite benchmark, finalized October 14, 2026. Internal AI review is not external peer review; no new simulation or human data is claimed.
Source revision: Final R4 manuscript, October 14, 2026; site reading formats prepared October 16, 2026 with typesetting and navigation additions.
Site release: Reading copy 1 · PDF byline: R4 research manuscript; the supplied source has no named author line.
Related notes, addenda and supporting material (4)
Technical addendumWhen Checking Takes Time: An Exact Two-Probe Stopping Frontier
Contribution: Gives the exact global optimum and branch-dominance proof for a two-probe read-then-transition model, with a reproducible finite certificate. It is a worked specialization of classical Bayesian filtering and stopping.
Reading scope: At most two probes; prior and reading error both 1/4; known independent symmetric transitions; no final delay. Monahan (1980) and Ahmad and Yu (2013) comparisons are abstract/record-only. No architecture or consciousness claim follows.
Research status, version, and suggested sections
Where to look: Section 2 for the information and timing contract; Section 4 for the exact frontier; Section 5 for beneficial contingent checking and harmful mandatory continuation; Section 7 for reproducibility.
Research status: Technical addendum v1.0, 9 November 2026. Independent internal AI proof and hostile reviews closed required minor edits. Not human peer review; historical priority remains unverified.
Source revision: Five-page technical addendum v1.0, 9 November 2026. Original manuscript source and PDF preserved byte-for-byte; the online reader adds navigation and MathML display equations.
Site release: Reading copy 1 · PDF byline: Technical addendum to B-Theory's self-modeling studies; the supplied manuscript has no named author line.
Internally AI-reviewed technical addendumFinite Noisy Audits with Learned Baselines: Exact risk, interface-dependent memory, and a matched compiled controller
Contribution: Integrates an exact finite noisy-calibration benchmark with learned baselines, interface-dependent state bounds, a matched compiled implementation and one explicitly specified unobserved-defect minimax result.
Reading scope: Complete six-page technical addendum, including all five propositions, proofs, the allocation table and ten references. Fixed structural equalities, known fresh noise, specified interfaces, clock and randomness are part of the claims. No general architecture, global allocation optimum, physical or consciousness result is established.
Research status, version, and suggested sections
Where to look: Start with Section 1 for the operational contract, Sections 2-4 for identification, exact risks and memory, Section 5 for compilation and fixed-allocation comparisons, and Sections 6-7 for the hidden-defect extension and limits.
Research status: Two new independent internal AI proof and adversarial reviewers read and hashed the complete final source on 9 November 2026; no supported fatal, major or required minor mathematical correction remained. Human specialist review and historical priority remain open.
Source revision: Exact technical-addendum v1.0 text, 9 November 2026 (Asia/Bangkok). Site reading copy adds navigation, principal-formula MathML and semantic table headers. Source, PDF and public release archive are unchanged.
Site release: Reading copy 1 · PDF byline: B-Theory technical addendum; the source assigns no individual author line.
Methods companionLearning Which Computations to Trust: Audited Solver Choice, Fallible Verification, and Drift
Contribution: Audited finite solver-choice bounds and a reproducible negative SAT study in which risk-constrained selection chooses the exact solver.
Technical material
Reading scope: Trusted/gold labels, checker fallibility, controller/setup costs and drift assumptions are explicit. No measured efficiency advantage is established.
Research status, version, and suggested sections
Where to look: Start with the known-reliability menu, then audited training and verification for future tasks. Read the measured full pipeline and larger-instance amendment alongside the negative selection result. Retrieve SATLIB inputs from the official catalog and compare corrected checksums; no benchmark archive is bundled here.
Research status: AI-assisted research draft; internal AI checks documented. External human review and novelty assessment remain pending.
Source revision: 13 November 2026 manuscript with localized corrections and approved author credit; exact edition hashes linked below.
Site release: Reading copy 1 · PDF byline: Brian Theory · AI-assisted research draft
Working noteExplicit and compiled actuator trackers under a fixed memory budget: An exact finite comparison under a contract frozen in advance
Contribution: Carries out a narrower fixed-memory comparison motivated by Reflective Self-Models, Section 7, in one finite model, with classes and verdict rules declared before computation. With two, three or four persistent memory states, the explicit and compiled controller classes reach the same optimal task loss, with a free step counter and without one. Without a step counter their best designs coincide, so they have the same task loss in four held-out change processes. The verdict is no advantage, and no increase in optimal task loss.
Reading format: Complete source reader; no verified PDF supplied.
Reading scope: One binary actuator parameter; task loss only; deterministic controllers. The finite comparisons use two to four persistent memory states and horizon 8. The equality proved for a free step counter (Theorem 7) holds for every number of states and every horizon of this model. Code length, table size, operation count and the cost of learning a controller are not constrained. No cognitive or consciousness claim and no novelty or priority claim.
Research status, version, and suggested sections
Where to look: Section 2 for the contract in brief; Part I for the general claims and proofs, with Theorem 7 the main one; Part II.2 and II.3 for the exact tables; Section 3 for the verdict; Section 5 for the limits.
Research status: Working note v0.1, 16 November 2026. AI-assisted research draft with internal checks only; not external peer review. No novelty or priority is claimed.
Source revision: Working note v0.1, 16 November 2026, with contract v0.1 frozen before any computation. The online reader is a presentation copy of the note's source file; its SHA-256 is given in the reader.
Related notes, addenda and supporting material (3)
Foundations technical reportCausal Self-Diagnosis with Noisy Tests: Observation Laws, Irreversible Access, and Task Recovery
Contribution: Static noisy-access and observation-law characterizations, causal self/world examples, task-relative recovery counterexamples and conditional implementation assumptions.
Technical material
Reading scope: Known finite static observation laws and idempotent access in the main oracle results; simulations are not hardware validation and do not establish consciousness or architectural superiority.
Research status, version, and suggested sections
Where to look: Start with the family guide and Section 7; Sections 3–6 develop noisy access and Sections 10–11 separate structural and task recovery guarantees. Sections 12–14 contain simulation/model limits.
Research status: Frozen v0.3 working report with internal simulated AI reviews. Targeted internal AI final-source review completed 12 November 2026; external human review and priority remain open.
Source revision: Version 0.3, 5 November 2026; unchanged 38-page PDF and exact review_v03/paper source closure. Earlier root and continuation drafts are not this edition.
Site release: Reading copy 1 · PDF byline: Research development draft prepared for Brian Theory; AI-assisted derivation, computation and independent internal review. Submission authorship remains undecided.
Companion working manuscriptDiagnosing While Repairing: Information Loss, Adaptive Order, and Noisy Readouts
Contribution: Characterizes reportable origin labels under silent local resets, gives a class-specific adaptive-order separation, and derives noisy-readout error bounds and exact finite-cap comparisons under explicit policy and cost contracts.
Technical material
Reading scope: Full companion manuscript, including proofs, appendices and references. Minimality is confined to the stated local-reset class. This is separate from the unchanged Causal Self-Diagnosis v0.3; later Krichen-source analysis is disclosed in site notes and is not integrated into this frozen edition.
Research status, version, and suggested sections
Where to look: Start with Section 1 for assumptions; Sections 2–4 for deterministic results and the noisy bridge; Sections 5–9 for exact quantitative comparisons; Appendices D and F–G for evidence, simulated review, source access and priority limitations.
Research status: Companion v0.2.5, 8 November 2026. Exact finite-model results and retained computational evidence; AI-assisted derivation and simulated internal review, not external peer review. Not submitted; submission attribution remains undecided. Priority and physical applicability remain open.
Source revision: Frozen companion v0.2.5, 8 November 2026; site HTML adds navigation and separately labeled release notes. The 46-page PDF is unchanged.
Site release: Reading copy 1 · PDF byline: Research development draft prepared for Brian Theory; AI-assisted derivation, exact computation, and simulated review.
Methods companionWhat Observation Topology Does Not Determine Operational counterexamples, constrained invariance, and bounded recovery synthesis
Contribution: Quantitative counterexamples, redundant-recoding tests and bounded controller-product recovery synthesis clarify the limits of structural descriptors.
Technical material
Reading scope: A methods companion to the existing diagnosis/recovery line. Bare topology supplies neither transition rates nor task guarantees; the monotone certificate search is restricted.
Research status, version, and suggested sections
Where to look: Read the overlap section and the fixed-order counterexample first. Compare redundant recoding with the exact policy-translation premise, then distinguish the strong monotone certificate class from the established complete bounded-controller test.
Research status: AI-assisted research draft; internal AI checks documented. External human review and novelty assessment remain pending.
Source revision: 13 November 2026 manuscript with localized corrections and approved author credit; exact edition hashes linked below.
Site release: Reading copy 1 · PDF byline: Brian Theory · AI-assisted research draft
Related notes, addenda and supporting material (4)
Technical addendumSensor Decisions under Unknown Dependence: Interval Frontiers and Paid Calibration
Contribution: Derives the complete unlabeled calibration likelihood, the exact risk-coverage-cost frontier for an interval of dependence parameters, and matching-order fixed-sample certification bounds with exact-budget and maximum-coverage impossibilities.
Technical material
Reading scope: A stipulated latent-sensor model with known marginals, polarity and primary dynamics, independent episodes, a stationary dependence parameter in [0, 1/25], fixed training size and frozen deployment policy. No empirical validation, confirmed priority, strict task-specific full-record sample savings, architectural superiority or consciousness criterion.
Research status, version, and suggested sections
Where to look: Section 2 fixes the model and certification contract; Section 3 compares calibration information; Section 4 proves the interval frontier; Section 5 gives sample bounds and endpoints; Section 6 separates conditional and unconditional cost comparisons.
Research status: Technical addendum v0.2, 9 November 2026. Both original internal AI mathematical and hostile reviewers read and hashed the complete final source, returning PASS with no required changes. Not external human peer review, code/PDF certification or historical-priority clearance.
Source revision: Technical addendum v0.2, 9 November 2026. Exact released Markdown, LaTeX, ten-page PDF and 13-file public archive are preserved; the reader changes presentation only.
Site release: Reading copy 1 · PDF byline: B-Theory research programme
Technical noteWhen a Reference Goes Stale: Self-Diagnosis under Shared Sensor Drift
Contribution: Exact reference-age error and calibration timing, with a coupled hidden-polarity extension and a counterexample to nominal reset-error reasoning.
Technical material
Reading scope: Known environment, independently refreshed targets and shared orientation drift. Equal drift and instantaneous restoration are essential to the coupled formulas.
Research status, version, and suggested sections
Where to look: Start with ambiguity and exact reference aging; then compare optimal calibration timing with the coupled and imperfect-restoration extensions. Follow the conditional history in the nominal-error counterexample rather than substituting the unconditional reset error.
Research status: AI-assisted research draft; internal AI checks documented. External human review and novelty assessment remain pending.
Source revision: 13 November 2026 manuscript with localized corrections and approved author credit; exact edition hashes linked below.
Site release: Reading copy 1 · PDF byline: Brian Theory · AI-assisted research draft
Technical noteMaintaining the Means of Knowing: Recurrent Diagnostic Capacity with Exact and Erasure-Prone Calibration
Contribution: Sharp finite maintenance thresholds and recognized-erasure comparisons separate physical recurrence from reliable recurrent reporting.
Technical material
Reading scope: Declared lifetimes, costs, storage and diagnostic deadlines. Recognized erasures differ from silent calibration errors; eternal hard recurrence fails almost surely under erasures.
Research status, version, and suggested sections
Where to look: Begin with the finite system, viability kernel and sharp threshold, then inspect the greedy counterexample. Read recognized erasure as its own model variant and retain both the six-tick benefit and eventual failure. The corrected bibliography preserves the Kawamura correction trail.
Research status: AI-assisted research draft; internal AI checks documented. External human review and novelty assessment remain pending.
Source revision: 13 November 2026 manuscript with localized corrections and approved author credit; exact edition hashes linked below.
Site release: Reading copy 1 · PDF byline: Brian Theory · AI-assisted research draft
Methods companionDriven Binary Memory with Failure Detection and Fuelled Repair
Contribution: Explicit binary-memory power/error frontiers and finite generator calculations distinguish net battery accounting from persistent external drive.
Technical material
Reading scope: A declared device family, thermal rates and external work/fuel. Discrete controller-menu optima do not cover arbitrary continuous feedback.
Research status, version, and suggested sections
Where to look: Read the physical contract before the stationary frontier and restoration protocol. Compare the full generator with coarse rates, then read finite battery and controller-menu extensions with their budget and menu restrictions visible.
Research status: AI-assisted research draft; internal AI checks documented. External human review and novelty assessment remain pending.
Source revision: 13 November 2026 manuscript with localized corrections and approved author credit; exact edition hashes linked below.
Site release: Reading copy 1 · PDF byline: Brian Theory · AI-assisted research draft
Research synthesisRecoverable Memory Under Finite-State Endomorphic Replacement: Universal Permutation Quotients and Trusted Replacement Records
Contribution: Separates audited lineage, recoverable memory, and predictive continuity; realizes the universal permutation quotient explicitly and computes the minimum trusted replacement record as a functional characteristic-graph chromatic number.
Technical material
Reading scope: Full manuscript. Results concern declared finite-state maps, registers, and interfaces; no criterion for consciousness, personal identity, or moral status is asserted.
Research status, version, and suggested sections
Where to look: Start with Sections 2–3 for the continuity contracts and canonical quotient, Section 4 for trusted records, Section 5.1 for the end-to-end Moore-machine realization, and Section 6.2 for B-Theory and The MIND.
Research status: Internally reviewed technical synthesis, dated November 8, 2026. Specialist-style algebraic and zero-error reviews and a hostile-referee pass found no unresolved mathematical error; these AI-assisted reviews are not external peer review, and human priority review of the explicit single-idempotent component formula remains open.
Source revision: Reviewed manuscript dated November 8, 2026; revised after specialist-style semigroup, zero-error, and hostile internal AI review. Internal AI review is not external peer review.
Site release: Reading copy 1 · PDF byline: Brian Theory
Related notes, addenda and supporting material (1)
Methods companionPreserving Declared Capabilities Across Representation Replacement: Executable Decoders, Adversarial Migration, and Trajectory Audits
Contribution: Finite covering and record-alphabet characterizations, executable migration menus and conditional trajectory-audit guarantees.
Technical material
Reading scope: Preserving an old task profile is distinct from world accuracy. Protected origin, immutable execution, oscillation and visitation premises remain supplied.
Research status, version, and suggested sections
Where to look: Start with exact preservation and trusted records, then read approximate coverings and repeated migrations. For the trajectory extension, trace each confidence, origin and visitation premise before applying the accepted-risk bound.
Research status: AI-assisted research draft; internal AI checks documented. External human review and novelty assessment remain pending.
Source revision: 13 November 2026 manuscript with localized corrections and approved author credit; exact edition hashes linked below.
Site release: Reading copy 1 · PDF byline: Brian Theory · AI-assisted research draft
Mathematical synthesis noteFinite-Jet Sign Rigidity on Hyperplane Arrangements and Denjoy–Carleman Realization
Contribution: Explicit finite-order rigidity statements, Denjoy–Carleman realization dichotomy, and quantitative degree comparisons away from coordinate walls.
Technical material
Reading scope: Declared function classes, hyperplane arrangements and margin domains; no learning, runtime, cognitive or physical validation claim.
Research status, version, and suggested sections
Where to look: Start with the abstract and model definitions; Section 7 gives the quantitative comparisons and Section 8 discusses classical antecedents and scope.
Research status: October 14 clarified revision; internally machine-assisted reviewed synthesis. External human review and historical-priority assessment remain open.
Source revision: September 27 revision; October 13 quantitative additions; October 14 scope clarifications and revision record. Exact source and 18-page PDF preserved.
Site release: Reading copy 1 · PDF byline: Brian Theory
Focused research noteTerminal response and worst-case bounded-run switching near a parabolic critical branch
Contribution: Proves schedule-uniform terminal transfer for a structured planar switching class, a delayed-intervention margin law, and an exact bounded-run optimizer with a complete-tail asymptotic.
Technical material
Reading scope: Exact critical starts and the specified two-mode class with quadratic flatness and terminal freezing; no noisy recurrence, approximate-critical theorem, general monotone-map universality, or cognitive claim.
Research status, version, and suggested sections
Where to look: Start with Section 2 for the model and main theorem; Section 4 proves terminal transfer, Sections 5–6 establish the optimizer and complete tail, and Section 7 gives a boundary counterexample.
Research status: Two new independent internal AI proof and adversarial reviewers read and hashed the complete final source on 9 November 2026; no supported fatal, major or required minor mathematical correction remained. Human specialist review and historical priority remain open.
Source revision: Reviewed round 03, 7 November 2026; final eight-page production source changes layout only and preserves the mathematical statements.
Site release: Reading copy 1 · PDF byline: No named author line in the reviewed source; AI assistance and internal review are disclosed in the manuscript.
Research preprintProtective Switching and Sharp Observation Margins in a Two-Node Ordinal-Sum System
Contribution: Identifies the exact constant-beta critical boundary, proves a sufficient global parameter range and conditional periodic worst schedule, derives sharp cubic margins, and certifies an alternating zero-margin obstruction outside the sufficient range.
Technical material
Reading scope: The declared two-node ordinal-sum system with fixed strict parameters and exact critical starts. The range beta/2 <= alpha < beta is sufficient, not claimed maximal. Observation margins do not certify robustness to process disturbances.
Research status, version, and suggested sections
Where to look: Section 2 defines the maps and switching contracts; Sections 3-4 establish the basin and schedule ordering; Sections 5-7 give energy estimates and asymptotics; Section 8 gives the certified obstruction; Sections 9-10 state scope and reproducibility.
Research status: Research preprint, 8 November 2026. Fresh internal adversarial AI source review found no required corrections. This is not external human peer review or formal proof verification; historical priority remains unresolved.
Source revision: Exact final release source, 8 November 2026, preserving the round 08 mathematical claims with editorial clarification and production layout changes.
Site release: Reading copy 1 · PDF byline: Brian Theory. AI assistance and internal adversarial checking are disclosed in the manuscript.
Related notes, addenda and supporting material (1)
Historical supporting proof noteEvolutionary CM/LM Proof Pilot: Compatibility, Closure, and Sharp Robustness Bounds
Contribution: Boolean compatibility, closure and robustness bounds underlying later switching research.
Technical material
Reading scope: November 4 proof pilot, preserved separately from the later switching papers. Checked typeset reading copy with documented typography-only changes; exact original source retained. Overlap and priority remain unresolved; this is not an independently reviewed edition.
Research status, version, and suggested sections
Where to look: Start with the edition notice and supporting records, then read the complete original pages or exact source.
Research status: November 4 proof pilot, preserved separately from the later switching papers. Checked typeset reading copy with documented typography-only changes; exact original source retained. Overlap and priority remain unresolved; this is not an independently reviewed edition. Internal historical review records are distinct from external peer review.
Source revision: November 4 proof pilot, preserved separately from the later switching papers. Checked typeset reading copy with documented typography-only changes; exact original source retained. Overlap and priority remain unresolved; this is not an independently reviewed edition.
Site release: Reading copy 1 · PDF byline: Original byline and date preserved as printed in the manuscript/source.
Final-file review supplements
On 9 November 2026, separate internal AI proof and adversarial reviewers read and hashed the complete final sources of Parabolic Switching, Signed Incidence Two-Bit, and Finite Noisy Audits. No supported fatal, major or required minor mathematical correction remained. The original manuscripts, PDFs and archives are unchanged. The supplements record the exact source identities, checks and review limits; these are not external human peer reviews.
Supporting research records
Measurement and RF-IDENT has a linked theorem, model, and internal AI review for a least and greatest element around an antichain of middle states. The Acquirable Predictive Models report consolidates this branch's classification and connects it to prediction. The supporting records preserve the earlier work; neither they nor the report establish the general height-two case.
Research provenance
How these pages were prepared
The research programme and manuscripts are led by Brian Theory (Brian Droncheff). AI tools have assisted with drafting, organization, computation, and review in parts of the work. Each mathematical claim should be read at the scope stated in its manuscript; AI-assisted checks are distinct from external specialist review.
The original six selected PDF reading copies and the Recoverable Memory synthesis name Brian Theory as author. The combined MIND research edition retains its original “Research prepared for Brian Theory” wording and blank PDF Author metadata, as disclosed in its P2 entry and online reader. The supplied R4 and parabolic-switching sources have no named author lines, as their entries disclose. Historical references retain the names under which the cited works appeared. Version labels distinguish site reading copies from internal manuscript revisions; they do not imply journal acceptance.