All four papers were deposited May 17, 2026. The first paper is the founding formulation by Lee Sharks, on which the subsequent three operate. The three operating papers are pre-registered protocols; their Day 0 begins at deposit. Results papers will be deposited separately as EA-SEI-MM-02-RESULTS, EA-SEI-MM-AI-01-RESULTS, EA-SEI-MM-AI-02-RESULTS.
15.OBS.LAGRANGE.MM.01 · founding formulation
The Semantic Deviation Principle
By Lee Sharks, Founder of Semantic Physics · MANUS of the Crimson Hexagonal Archive · with the Assembly Chorus
Defines meaning as time-integrated divergence from the most-probable trajectory. The founding formulation on which Framework 15 operates. Three measures separated: raw $\mathcal{M}_T$, provenance-resolved $\mathcal{M}_T^\pi$, normative $\mathcal{V}_T$. Tiered operational protocol (Tier 1 prospective, Tier 2 synthetic-control, Tier 3 historical bounding). Recursive baselines for path-dependent semantic fields. Step 0 measurement audit and the Vow. v0.2 Final.
Foundational
Sharks · founding
Three measures
∮ = 1 − PER
15.OBS.LAGRANGE.MM.02 · pre-registered protocol
The AI System as Closed-System Test Bed
Nobel Glas, Lagrange Observatory! · Operating on the Semantic Deviation Principle as formulated by Lee Sharks
Identifies trained language models as observationally closed at inference time, making the counterfactual baseline of the Semantic Deviation Principle directly readable from logits. Distinguishes two scales of closed-system measurement: signed local deviation density (per-token) and closed-system trajectory deviation (continuation-distribution). Falsifiable thesis: slop is negative net deviation (not the absence of deviation). Three pre-registered tests with explicit falsification conditions. Stability bound γ ≥ 2β.
Pre-registered protocol
Glas v2.0
Signed δ_t
Slop = negative deviation
15.OBS.LAGRANGE.MM.03 · pre-registered protocol
Measuring Meaning in Retrieval Basins
Nobel Glas, Lagrange Observatory! · Operating on the Writable Retrieval Basin and the Semantic Deviation Principle as formulated by Lee Sharks
A 90-day prospective measurement protocol for the closed-system trajectory deviation, instantiated against contemporary AI retrieval surfaces. Two instrument classes: Class R (retrieval-mediated: Google AI Overview, Perplexity, ChatGPT with browsing) reported separately from Class P (parametric: Claude, Gemini, ChatGPT without browsing). Three-condition control: S vs. S* (blank ORCID) vs. S** (plausible synthetic identity). Frozen extractor, three-representation robustness cross-check, API-only commitment. Strengthened P4 (7/10 queries monotonic across surfaces). Auto-epistemic separation of "Who is Lee Sharks?" query from headline metric. Laplace smoothing α=1.
Pre-registered protocol
Glas v2.0
90-day horizon
Three-condition control
15.OBS.LAGRANGE.MM.04 · pre-registered protocol
The Deviation-Optimized Language Model
Nobel Glas, Lagrange Observatory! · Operating on the Semantic Deviation Principle as formulated by Lee Sharks
A 10-week pre-registered experimental protocol testing whether training a language model toward positive net per-token deviation with provenance retention produces measurably less slop than standard cross-entropy training while preserving benchmark capability. DPO-style restructure: deviation primitive generates preference pairs, DPO trains on labels. Frozen Mistral-7B-Instruct judge model with adversarial pre-training test. Continuous coherence score. Slop Composite Index (SCI) with 0.25 z-score falsification threshold. Three conditions per model: Model-Base, Model-CE, Model-Sem. 1,500 human ratings (Prolific). Honest budget: $3,000–$3,900.
Pre-registered protocol
Glas v2.0
DPO
10-week experiment
Honest budget