Research for intelligence that persists, acts, and proves what changed.
Language is not completion. Glyphd Labs studies the state, address, authority, representation, and interface layers that let humans and models share a durable world.
Evidence states advance only when a versioned receipt supports the move.
GL–2026.01
31FULL PAPERS- 21
- field notes
- 26
- instruments
- 19
- test contracts
“Generating language is not the same thing as doing what the language says.”
Pages do not begin at PASS. They earn stronger language as mechanisms survive experiments and return inspectable receipts.
One architecture, multiple frontiers.
The work is organized by responsibility. Truth is not the lens. Representation is not authority. Semantic resemblance is not exact identity. A model proposal is not an accepted act.
Continuity & Canon
Memory, authority, provenance, identity, and reconstruction that do not depend on a model remembering correctly.
Open program →02Address & Locality
Exact and typed identities, foveated retrieval, locality, and reusable computation that preserve proof.
Open program →03Actors & Agents
Persistent actors, scoped agents, Work Orders, Returns, verification, and human acceptance.
Open program →04Regenerative Objects
Objects, video, temporal media, and model flights routed through exact, inspectable representations.
Open program →05Compact Intelligence
4B-class models strengthened by contracts, surfaces, schemas, deterministic runtimes, and evidence gates.
Open program →06Manifested Interfaces
Workpages, spatial intent, visual-world-first software, referential return, and artifact-first operation.
Open program →Mechanisms before marketing.
Generating Language Is Not the Same as Doing What the Language Says
The manifestation gap and the case for stateful latent objects in general AI assistance
Language models can describe a plan, simulate the voice of an expert, enumerate the steps of a build, and announce that work is complete without any necessary change occurring in the world that the language des…
The Future Is Foveated
Deterministic addressing, procedural landmarks, and evidence-bound navigation
Large information systems are typically navigated by names, search results, folders, timelines, and lists. These are effective but force every return to begin as a query or a reconstruction of hierarchy. Human…
GlyphCAS
Exact content, typed semantic identity, and evidence-preserving reuse for language-model systems
Conventional content-addressable storage is exact by design: identical bytes receive identical content identities; different bytes do not. Language-model systems often want a second property: repeated meanings,…
Actor Before Agent
Persistent identities, temporary inference, and evidence-bound World Patches
Most “agent” architectures begin with a running model and ask how to give it memory, personality, tools, and autonomy. That order makes identity dependent on continuous inference and encourages systems to confu…
A Provenance-Routed Regenerative Object Codec
The fold/2 P/D/M/R/G representation system with exact proof and fallback
General-purpose compression assumes that an object is primarily a byte sequence. Many modern objects also have recoverable context: a prior version, a shared dictionary, an exact source, a deterministic generat…
Small Models, Strong Contracts
EL-JEFFE 4B as a local-first executive intelligence with deterministic authority and receipts
Compact language models are increasingly capable, but a small local model is often judged against a cloud model on the cloud model's preferred task: absorb a large context, reason freely, write a complete answe…
Status is earned, not styled.
The institute separates reproduced results, verified mechanisms, implementation state, source review, specifications, hypotheses, inconclusive evidence, failed runs, and experiments awaiting local hardware.
Inspect the evidence ledger