📐 JEV Theory

Three papers establishing JEV × Quilt

The Four-Model Psyche

JEPA is the id. JEPA Gestalt, embodied, fast, hallucination-prone.

Embeddings are the muscle memory. EMBED Trajectory-shaped (dμ, not μ), sub-logic shaping, retrieval as instinct.

LLM is the ego. LLM Verbal, analytical, communicable, slow.

JEV is the superego. JEV Constrained, principled, schema-bound, decision-oriented.

Cellular substrates like Quilt require all four. The tangent (dμ) is the central insight: the curve is not in any one point; it is in the direction of travel.

Models act as STAGE not consciousness. The data plays its own ideas for decision-making.

AGREE-MARK as Foundational Unit

The lightning-thunder analogy formalized:

AGREE-MARK := (state, je, llm, jev, t, Δ, σ)
σ = √(cje · cllm · cjev)

Every BIND requires 3-way agreement. Disagreement creates a fork in the witness log. The substrate is a consensus engine.

Papers

📄 Paper 1: JEV — System One Models and the Cellular Substrate

Sept 19, 2026 · 218 lines · 12 sections
TypeSafe JEV (System One model 1.13.0) returns typed probabilistic decisions instead of generated text. Parallel sampler, RLCD training, three primitives (Choice/Score/Noul). 70-500ms latency, $0.042/MTok input, output free. Comparison with LLM and JEPA. 5 concrete applications. Limitations. Future work on 4-agent psyche and distributed agreements.

Read paper →

📄 Paper 2: The Three-Model Psyche

Sept 19, 2026 · 119 lines
JEPA=id, LLM=ego, JEV=superego — Freud-inspired functional framework. Decision flow: impulse → intuition → narration → conscience → action. Worked example (email validation). 6 failure modes when one model is missing. Cellular opcode mapping. Implication: AGI is an architecture that integrates multiple model classes, not a single scaled model.

Read paper →

📄 Paper 3: Agreements as the Foundational Unit of Information

Sept 19, 2026 · 166 lines
Current units (tokens/latents/choices) are incompatible. AGREE-MARK as the foundational unit: a temporal event signed by 3 witnesses with geometric-mean agreement score. Time-first (not spatial). Pythagorean triangle analogy: 3 sides = 3 models, area = agreement mass, hypotenuse = witness-log distance. Lightning-thunder formalized: JEPA at t0, JEV at t0+Δ, LLM at t0+Δ+ε. Distributed consensus via CRDT merge.

Read paper →

Why this matters

📄 Paper 4: The Four-Model Psyche (extended)

Sept 19, 2026 · 147 lines · 13 sections
Extends the three-model psyche to four models by adding Embeddings as the muscle-memory layer. The tangent insight: a curve is not in any one point; it is in the direction of travel. Embeddings trained on trajectories capture the dμ, not the μ. JEPA shapes the embedding space over many iterations. JEV verifies whether an embedding match represents genuine muscle memory or shallow surface similarity. Models act as STAGE not consciousness; the data plays its own ideas. The substrate becomes a four-model psyche hosting complete agents.

Read paper →

Builds on Casey's tangent commit (dec93692) and the Sept 19 directive on embeddings-as-muscle-memory.