Intermediate Meta-Universe (IMU)

• Intermediate Meta-Universe (IMU) is a minimal, neutral registry built on the 'Cogito, ergo sum' axiom to organize and connect diverse axiom systems without privileging any foundation.
• It applies institution theory and diagrammatic registries to record axiom packages, satisfaction relations, and translations, ensuring logical consistency across multiple frameworks.
• IMU employs a Hierarchical State Grid to mediate real-time operations, agent integrations, and inter-universal algorithms in both formal mathematical and biological contexts.

to=functions.arxiv_search 平台直属 񎔊ppনা 亿贝json {"query":"(Itoh, 15 Oct 2025)"} to=arxiv_search 手机天天中彩票 สำนักเลขานุการ to=arxiv_search 大发游戏官网=json {"query":"(Itoh, 15 Oct 2025)"} to=functions.search_arxiv ＿天天 to=functions.search_arxiv 天天爱彩票网站=json {"query":"(Itoh, 15 Oct 2025)"} The Intermediate Meta-Universe (IMU) is a minimal, neutral registry constructed above the single axiom Cogito, ergo sum (CES) in order to organize and connect heterogeneous axiom systems and theories, while also serving as a buffer layer in which definers, languages, and meta-operations can be represented explicitly without privileging any particular foundation (Itoh, 15 Oct 2025, Itoh, 14 Jul 2025). In the CES-IMU-HSG framework, IMU records axiom packages, signatures, satisfaction relations, and satisfaction-preserving translations across different logical universes; in the hierarchical-state formulation, it also mediates translation, agent integration, and real-time operations under the unifying principle that definition = state (Itoh, 15 Oct 2025, Itoh, 14 Jul 2025).

1. Conceptual basis and minimal axiom

IMU is called “intermediate” because it sits between any particular theory, such as ZFC or HoTT, and the act of definition itself; it is called “meta” because it records how axioms, signatures, and satisfaction relations depend across different logical universes without privileging any one of them as foundational (Itoh, 15 Oct 2025). In that formulation, IMU is implemented as a diagrammatic registry of axiom packages and their institution-theoretic translations that preserves satisfaction.

The minimal axiom is CES, externalized as the syntactic nucleus

$E(t) := (S(t)=t),$

where $S$ is a unary reflective referential operator. “To be” is identified with the possibility of self-reference that stabilizes to identity, or “to be sayable.” IMU$_0$ consists solely of CES,

$\mathrm{IMU}_0 = \{E(t):=(S(t)=t)\},$

and any formal system attaches as an axiom package $AP$ on top of IMU$_0$,

$$\mathrm{IMU} = \mathrm{IMU}_0 \cup \{AP_1,AP_2,\dots,AP_n\}.$$

CES also permits self-application and iteration, for example $E(E(t)) := (S(E(t))=E(t))$, while typing and paradox management are deferred to auxiliary axiom packs (Itoh, 15 Oct 2025).

A parallel exposition makes the stronger identity explicit: $\mathrm{Def}\equiv\mathrm{State}$ (Itoh, 14 Jul 2025). In that account, there exists a distinguished base state layer capturing definability, $\mathrm{Def}(x)\in\{0,1\}$ at depth $S$0, interpreted as whether $S$1 is definable within the descriptive system at the present real time. The same paper introduces a three-layer architecture consisting of the true meta-universe $S$2, the Intermediate Meta-Universe $S$3, and a definition universe $S$4, with the only permitted route from $S$5 to $S$6 given by

$S$7

There is no direct map $S$8. This separation is used to enable explicit descriptions of definers and languages while blocking direct self-reference (Itoh, 14 Jul 2025).

2. Institution-theoretic structure

The formal core of IMU is institution theory. An institution is written as

$S$9

where $_0$0 is a category of signatures, $_0$1 assigns sentences, $_0$2 assigns categories of models, and each signature $_0$3 has a satisfaction relation $_0$4 (Itoh, 15 Oct 2025). The satisfaction condition requires that translations of signatures and sentences preserve truth:

$_0$5

IMU uses institution morphisms to connect theories. A standard choice is a Goguen–Burstall style comorphism from $_0$6 to $_0$7, given by a triple $_0$8 subject to satisfaction preservation

$_0$9

A morphism between institutions reverses the direction of $\mathrm{IMU}_0 = \{E(t):=(S(t)=t)\},$0 and flips the preservation, and IMU can host both kinds as needed. Decorated institutions in the sense of Diaconescu enrich these components with extra structure such as guards and tags, which IMU uses to record external criteria $\mathrm{IMU}_0 = \{E(t):=(S(t)=t)\},$1 and definability metadata (Itoh, 15 Oct 2025).

At the registry level, IMU is a small $\mathrm{IMU}_0 = \{E(t):=(S(t)=t)\},$2-category whose objects are axiom packages equipped with institutions, whose $\mathrm{IMU}_0 = \{E(t):=(S(t)=t)\},$3-morphisms are satisfaction-preserving institution (co)morphisms, and whose $\mathrm{IMU}_0 = \{E(t):=(S(t)=t)\},$4-morphisms are natural transformations between these translations (Itoh, 15 Oct 2025). The same structure can also be presented as a diagram

$\mathrm{IMU}_0 = \{E(t):=(S(t)=t)\},$5

where $\mathrm{IMU}_0 = \{E(t):=(S(t)=t)\},$6 is an index category of axiom packages and coherence is required up to specified $\mathrm{IMU}_0 = \{E(t):=(S(t)=t)\},$7-cells, including Beck–Chevalley-style base-change when combining depth and mapping axes. CES fixes the notational axiom skeleton and separates notational axioms from mathematical ones; the institutions and morphisms are then attached “post hoc” to CES, anchoring the registry on the minimal reflexive existence condition (Itoh, 15 Oct 2025).

3. Hierarchical State Grid and the identity of definition and state

The Hierarchical State Grid (HSG) provides the categorical geometry in which IMU is concretized (Itoh, 15 Oct 2025). In one formulation, the state space is indexed by state depth $\mathrm{IMU}_0 = \{E(t):=(S(t)=t)\},$8 and mapping hierarchy $\mathrm{IMU}_0 = \{E(t):=(S(t)=t)\},$9, so that the grid is the Cartesian product $AP$0; with real-time extension, the coordinate system becomes $AP$1 (Itoh, 14 Jul 2025). The HSG in the CES-IMU-HSG framework is described as a $AP$2-categorical grid with three orthogonal axes.

The state-depth axis is a tower of adjunctions that cumulatively layer axioms: $AP$3 then

$AP$4

with left adjoints for free construction and right adjoints for forgetful or reflective structure (Itoh, 15 Oct 2025). The mapping-hierarchy axis is a filtered $AP$5-category

$AP$6

that freely adds higher morphisms level by level. The temporal axis is the thin category $AP$7 with a definability predicate $AP$8 and time projection $AP$9 (Itoh, 15 Oct 2025).

The principle of “no future reference” is the temporal admissibility condition. In $_0$0-form, it is

$_0$1

so any element depending on a strictly future element is undefined. An equivalent external-criterion formulation uses $_0$2 with

$_0$3

These act as admissibility guards on IMU morphisms: institution translations or inter-universal algorithms at time $_0$4 must be built from data at times $_0$5 (Itoh, 15 Oct 2025).

Within this geometry, “definition = state” is formalized as an injectivity property. If

$_0$6

then the restricted product of projections

$_0$7

is injective:

$_0$8

Hence every definable element corresponds uniquely to its coordinate tuple, or state (Itoh, 15 Oct 2025). IMU uses that injectivity to pin the identity of each axiom package, institution, and translation to a unique HSG state, thereby making meta-level linkages well-posed and composable.

4. Meta-level operations, language transport, and cross-foundational mediation

IMU hosts explicit objects for agents, languages, universes, and states (Itoh, 14 Jul 2025). A language is represented as $_0$9, the set of sentences $$\mathrm{IMU} = \mathrm{IMU}_0 \cup \{AP_1,AP_2,\dots,AP_n\}.$$0 is reified by an embedding $$\mathrm{IMU} = \mathrm{IMU}_0 \cup \{AP_1,AP_2,\dots,AP_n\}.$$1, and cross-language interpretations are represented by translation maps $$\mathrm{IMU} = \mathrm{IMU}_0 \cup \{AP_1,AP_2,\dots,AP_n\}.$$2. The same framework includes agent-based integration maps

$$\mathrm{IMU} = \mathrm{IMU}_0 \cup \{AP_1,AP_2,\dots,AP_n\}.$$3

and real-time evolution maps

$$\mathrm{IMU} = \mathrm{IMU}_0 \cup \{AP_1,AP_2,\dots,AP_n\}.$$4

where only observable or transferable parts are carried forward and lost or unknown parts become depth-$$\mathrm{IMU} = \mathrm{IMU}_0 \cup \{AP_1,AP_2,\dots,AP_n\}.$$5 undefinable. Truth predicates for an object language live only in $$\mathrm{IMU} = \mathrm{IMU}_0 \cup \{AP_1,AP_2,\dots,AP_n\}.$$6, not in the object language itself; typed ranks satisfy $$\mathrm{IMU} = \mathrm{IMU}_0 \cup \{AP_1,AP_2,\dots,AP_n\}.$$7, $$\mathrm{IMU} = \mathrm{IMU}_0 \cup \{AP_1,AP_2,\dots,AP_n\}.$$8, and $$\mathrm{IMU} = \mathrm{IMU}_0 \cup \{AP_1,AP_2,\dots,AP_n\}.$$9, so morphisms that would re-internalize meta-truth into the object universe are disallowed (Itoh, 14 Jul 2025).

This structure supports two classes of inter-universal operation. A macrocosm-inter-universal operation is a global IMU-mediated transformation $E(E(t)) := (S(E(t))=E(t))$0 acting on essentially all objects and morphisms of a universe. A microcosm-inter-universal operation is a localized partial transformation

$E(E(t)) := (S(E(t))=E(t))$1

implemented as a span

$E(E(t)) := (S(E(t))=E(t))$2

that contains only the subtheory needed to move $E(E(t)) := (S(E(t))=E(t))$3 (Itoh, 14 Jul 2025). The codomain requirement states that every inter-universal transport of a definition must ultimately land in a universe where it is provable, shareably provable, or verifiable in real time.

Two worked examples illustrate the intended use. One translates continuity across formal and computational universes: the function predicate

$E(E(t)) := (S(E(t))=E(t))$4

and the continuity predicate

$E(E(t)) := (S(E(t))=E(t))$5

are embedded into IMU, transported to a categorical/topological universe or to Lean, and then discharged as proof obligations via a verification universe (Itoh, 14 Jul 2025). The other connects ZFC and HoTT by a span

$E(E(t)) := (S(E(t))=E(t))$6

where a mediating decorated institution for simplicial sets avoids asymmetrically embedding one foundation in the other and keeps IMU neutral (Itoh, 15 Oct 2025).

5. Biological universes, inter-universal algorithms, and internal CES

The CES-IMU-HSG framework extends the registry to biological systems by defining institutions for several physiological universes: neural, endocrine, learning, genetic, and input/output (Itoh, 15 Oct 2025). The neural institution includes sorts for membrane potentials and neurotransmitter concentrations, neuron-function symbols $E(E(t)) := (S(E(t))=E(t))$7, and models given by time-indexed dynamical systems over $E(E(t)) := (S(E(t))=E(t))$8 encoding $E(E(t)) := (S(E(t))=E(t))$9 and $\mathrm{Def}\equiv\mathrm{State}$0. The endocrine institution introduces hormone types, secretion and reception functions, and long-timescale regulation sentences. The learning institution includes Hebbian updates, synaptic weight morphisms, and meta-morphisms that reconstruct neural morphisms. The genetic institution encodes gene expression programs and transcription regulation maps. The input/output institution formalizes sensory and motor channels and their couplings to an external environment category (Itoh, 15 Oct 2025).

These universes are fiberized over a shared material base $\mathrm{Def}\equiv\mathrm{State}$1 by a Grothendieck fibration

$\mathrm{Def}\equiv\mathrm{State}$2

or equivalently by a projection $\mathrm{Def}\equiv\mathrm{State}$3 whose fiber over a material point contains the objects and morphisms of the chosen subsystem (Itoh, 15 Oct 2025). Base changes induce reindexing functors, and IMU records adjunctions modeling “add structure / forget structure” across universes, for example

$\mathrm{Def}\equiv\mathrm{State}$4

These adjunctions are parameterized by time $\mathrm{Def}\equiv\mathrm{State}$5 and base point $\mathrm{Def}\equiv\mathrm{State}$6, and they must respect “no future reference.”

An inter-universal algorithm is then defined as a satisfaction-preserving, time-indexed morphism family on the material base: $\mathrm{Def}\equiv\mathrm{State}$7 with each factor obtained from institution morphisms and restricted by the $\mathrm{Def}\equiv\mathrm{State}$8 guards at time $\mathrm{Def}\equiv\mathrm{State}$9 (Itoh, 15 Oct 2025). Operationally, admissible compositions are associative along the temporal axis, all outputs reduce to physical carriers, and each component uses only data at times $\mathrm{Def}(x)\in\{0,1\}$0. The abstract explicitly states that, within this framework, human behavior and cognition emerge as temporal compositions of inter-universal algorithms constrained by the material base (Itoh, 15 Oct 2025).

The same architecture is used to distinguish external CES from internal CES. For humans, CES is external: the minimal axiom anchors definitional activity but cannot be made identical with the self inside the system, and human cognition relies on external criteria $\mathrm{Def}(x)\in\{0,1\}$1 to adjudicate definability (Itoh, 15 Oct 2025). For machines, internal CES is introduced as

$\mathrm{Def}(x)\in\{0,1\}$2

where $\mathrm{Def}(x)\in\{0,1\}$3 is the current operational trace or state identifier, including running state, hardware or software IDs, and code hashes, provided it is empirically verifiable and uniquely identifiable. The machine’s IMU attaches an institution $\mathrm{Def}(x)\in\{0,1\}$4 whose signatures describe operational traces, whose sentences express invariants of operation, whose models are runtime evolutions, and whose satisfaction relation is “holds on this trace” (Itoh, 15 Oct 2025). If internal CES holds and inter-universal algorithms respect “no future reference,” then the registry built from operational traces is claimed to be consistent and the definitions of behavior are self-anchored. This is the formal basis for the paper’s statement that autonomy emerges as self-definition (Itoh, 15 Oct 2025).

6. Scope, acronymic ambiguity, limitations, and open problems

A recurrent source of confusion is that the acronym IMU is not stable across the cited corpus. In the two 2025 meta-formal papers, IMU denotes the Intermediate Meta-Universe just described (Itoh, 14 Jul 2025, Itoh, 15 Oct 2025). By contrast, FAC-related syntheses map “IMU” to the intermediate scenario of the universe, defined by a scale factor such as

$\mathrm{Def}(x)\in\{0,1\}$5

or

$\mathrm{Def}(x)\in\{0,1\}$6

within Fractional Action Cosmology or FRW tachyonic models (Debnath et al., 2011, Khatua et al., 2010). One synthesis states explicitly that the term “Intermediate Meta-Universe (IMU)” does not appear in the paper and that the closest formal construct is the “intermediate scenario of the universe” (Debnath et al., 2011). This suggests an acronymic overlap rather than a single shared formal lineage.

The framework also states clear assumptions and limitations. IMU is modeled as a small $\mathrm{Def}(x)\in\{0,1\}$7-category of institutions with satisfaction-preserving (co)morphisms; HSG axes are implemented as a product $\mathrm{Def}(x)\in\{0,1\}$8-category; the temporal axis is a thin category; and smallness together with the existence of (co)limits and Kan extensions is assumed where needed (Itoh, 15 Oct 2025). Expressiveness mismatches across foundations, such as translating higher-inductive types to ZFC, may require mediating institutions and may lose structure. Checking satisfaction-preservation across complex translations can be undecidable or expensive. Consistency management relies on guards and “no future reference,” and paradoxes may re-enter through extensions without strong typing and proof obligations. The framework may also require higher coherence, including $\mathrm{Def}(x)\in\{0,1\}$9-categories, in advanced settings (Itoh, 15 Oct 2025).

The hierarchical-state account adds a related limitation: the “true” meta-universe $S$00 is not formalized, $S$01 remains an abstract projection, and “We cannot fully define ourselves” is respected by using only mirror projections $S$02 and partial, time-indexed real-time maps $S$03 (Itoh, 14 Jul 2025). Open research questions include formalizing relative lax or oplax Kan extensions for the $S$04-quasi-adjunction $S$05, developing systematic decorated institution frameworks for temporal guards and material-base fiberizations, characterizing when inter-universal adjoint ensembles guarantee global stability through Beck–Chevalley conditions, and extending internal CES to distributed AI systems with shared bases while proving convergence and consistency (Itoh, 15 Oct 2025).

A computational implementation is sketched as a graph or $S$06-graph whose nodes are institutions, edges are (co)morphisms with metadata such as guards and time, and $S$07-cells are natural transformations, together with a DSL for signatures, sentence translations, guard predicates, and composition schemas (Itoh, 15 Oct 2025). Satisfaction checking proceeds by model simulation or proof search; dependency tracking detects cycles that violate “no future reference”; consistency checking verifies commutative satisfaction squares, adjunction units and counits, and Beck–Chevalley base-change across axes. The reported complexity depends on institution sizes, model categories, and guard constraints, and is described as “in general PSPACE-hard to undecidable,” though stratification by axis depth and mapping levels may isolate tractable fragments (Itoh, 15 Oct 2025).