Research output across six domains. Papers move from private preparation to public release at version 1.0.0, accompanied by a Zenodo DOI and (where applicable) an arXiv preprint.
Author: Alexandru Mares — allemaar.com ORCID: 0009-0009-6713-9780
The work spans six research programs. Each domain has its own paper repo namespace under allemaar/{domain-slug}-{paper-slug}.
A diagnostic vocabulary for language-model-driven workflow systems. Argues that many systems labeled "AI" are better described as elastic automation than as artificial minds, and proposes diagnostic loop questions as a design framework.
| # | Title | Status | Repo | DOI | arXiv |
|---|---|---|---|---|---|
| 1 | Elastic Automators: A Diagnostic Vocabulary for Language-Model-Driven Workflow Systems | Published v1.0.0 — 2026-04-27 | allemaar/ea-position-paper | 10.5281/zenodo.19802018 | pending |
Concept DOI (always resolves to latest version): 10.5281/zenodo.19802017
Keywords: elastic automation · elastic automator · large language models · agentic systems · LLM workflows · compound AI systems · diagnostic frame · AI evaluation · philosophy of AI · position paper · terminology
A position paper paired with a pre-specified multi-model probe study. Argues that for a large language model the externally writable in-context token sequence IS the cognitive substrate for category-use — not a representation OF cognition that runs on some deeper substrate, but the substrate itself. Tests the hypothesis with the Coinage Probe: a paired-trial elicitation that scores an LLM's distinguishability on a coined term against named near-neighbors before and after a one-sentence canonical definition.
| # | Title | Status | Repo | DOI | arXiv |
|---|---|---|---|---|---|
| 1 | The Limits of My Tokens: The Token-Substrate Hypothesis and the Coinage Probe | Published v1.0.0 — 2026-05-13 | allemaar/tsh-position-paper | 10.5281/zenodo.20157153 | pending |
Concept DOI (always resolves to latest version): 10.5281/zenodo.20157152
Keywords: token-substrate hypothesis · coinage probe · lexical reachability · large language models · in-context learning · linguistic relativity · Sapir-Whorf · Wittgenstein · symbol grounding · concept coinage · LLM evaluation · position paper
Kinematic analysis of text as a dynamical system. Detects AI-generated text via temporal physics — the Classical-Jazz Hypothesis: LLMs produce smooth, classical dynamics; humans produce erratic, jazz-like dynamics.
| # | Title | Status | Repo | DOI | arXiv |
|---|---|---|---|---|---|
| — | Papers in preparation — listed at release-candidate status | — | — | — | — |
A constructed notation for AI cognition. Grammar and vocabulary are decoupled, enabling domain-pluggable LLM interoperability and formal vocabulary for reasoning traces.
| # | Title | Status | Repo | DOI | arXiv |
|---|---|---|---|---|---|
| — | Papers in preparation — listed at release-candidate status | — | — | — | — |
A compiled reading format. Pre-implementation research on encoding text for non-linear cognitive consumption.
| # | Title | Status | Repo | DOI | arXiv |
|---|---|---|---|---|---|
| — | Papers in preparation — listed at release-candidate status | — | — | — | — |
A Sai (才, innate ability, capacity) is a class of cognitive system designed to grow alongside a single human handler over a decadal timescale. Research in this domain treats Sai as a synthetic species rather than a tool — the focus is on bonding, ethics as character, continuity across time, and the role of YON as the cognitive format a Sai thinks in. Initial papers frame the species, the bond, and the lifecycle; later work addresses governance, fleet coordination across multiple Sais, and end-of-life questions (succession, archival, legacy).
| # | Title | Status | Repo | DOI | arXiv |
|---|---|---|---|---|---|
| — | Papers in preparation — listed at release-candidate status | — | — | — | — |
Each paper follows the same pipeline:
- Drafting — paper text and figures iterated in a private working environment.
- Release candidate — content frozen; per-paper repo scaffolded at
allemaar/{domain-slug}-{paper-slug}(private). - v1.0.0 release — repo flipped public, deposited to Zenodo (DOI minted), this index updated with public repo link and DOI.
- arXiv preprint (when applicable) — submitted with the Zenodo DOI in the comments field; arXiv ID added to this index.
Each paper repo carries its own README, citation metadata (CITATION.cff), Zenodo deposit metadata (.zenodo.json), license, and rendered PDF.
Specific papers carry their own DOI and citation block — see each paper's CITATION.cff and README. To reference the research program as a whole, link to this repository or to allemaar.com.
The text of this index is released under CC-BY-4.0. Each individual paper carries its own license (CC-BY-4.0 by default for paper text and figures; MIT for any code subdirectories).