Research on AIBussin — AI applications, systems and books

Research: From Paper to Pipeline with AI

Fri, 28 Aug 2026 10:00:00 +0100

“Research is formalized curiosity. It is poking and prying with a purpose.”

Summary

There is still something powerful about picking up a research paper and asking:

What can I build from this?

That question turns reading into motion. A paper stops being a sealed object. It becomes a source of ideas, mechanisms, tests, questions, and possible systems.

But the modern version of this workflow is more disciplined than the old one.

Appendix 05: References: Foundational Research

Fri, 28 Aug 2026 10:00:00 +0100

    flowchart TD
    subgraph Research_Papers
        R1[Wei et al., 2022<br>Chain of Thought Prompting]
        R2[Gao et al., 2023<br>GRA Framework]
        R3[Zelikman et al., 2022<br>STaR]
        R4[Wu et al., 2024<br>xVerify]
    end

    subgraph Book_Chapters
        C4[Chapter 4: Beyond Prompting]
        C5[Chapter 5: Ghosted by the Machine]
        C9[Chapter 9: Council of Purpose]
        C10[Chapter 10: Second Pass]
        C13[Chapter 13: Machine Coding]
        C14[Chapter 14: MIRROR & Reflection]
        C15[Chapter 15: When It Doesn't Work]
        CApp[Appendix: Prompting Tools]
    end

    R1 --> C4
    R1 --> C5
    R1 --> C14

    R2 --> C9
    R2 --> C14
    R2 --> C15

    R3 --> C4
    R3 --> C5
    R3 --> C13

    R4 --> C10
    R4 --> C14
    R4 --> CApp

    style R1 fill:#e1f5fe,stroke:#333
    style R2 fill:#fff3e0,stroke:#333
    style R3 fill:#e8f5e9,stroke:#333
    style R4 fill:#fce4ec,stroke:#333
    style C4 fill:#f3e5f5,stroke:#333
    style C5 fill:#fff9c4,stroke:#333
    style C9 fill:#e0f7fa,stroke:#333
    style C10 fill:#c8e6c9,stroke:#333
    style C13 fill:#ffccbc,stroke:#333
    style C14 fill:#d1c4e9,stroke:#333
    style C15 fill:#b2dfdb,stroke:#333
    style CApp fill:#ffcdd2,stroke:#333

Wei et al., 2022. “Chain of Thought Prompting Elicits Reasoning in Large Language Models”
Relevance: Forms the backbone of your structured prompting techniques like MARS, GROWS, and CRITIC. CoT prompts are foundational for steering AI toward better reasoning.
Used in: Chapters 4 (Beyond Prompting), Chapter 5 (Ghosted by the Machine), Chapter 14 (Session Reflection)
https://arxiv.org/abs/2201.11903

Appendix 15: Paper Coverage Report: AI as a Co-Scientist

Fri, 28 Aug 2026 10:00:00 +0100

Based on:
van der Wal, R., et al. (2024). AI as a Co-Scientist: Hypothesis Generation and Evaluation with LLMs. arXiv:2502.18864v1

    flowchart TD
    subgraph Paper_Implementation
        P1[Hypothesis Generation<br>GenerationAgent + DSPy]
        P2[Self-Critique / Reflection<br>ReflectionAgent]
        P3[Ranking / Evaluation<br>ELO-style pairwise comparison]
        P4[Evolution & Grafting<br>EvolutionAgent]
        P5[Meta-Review / Summary<br>MetaReviewAgent]
        P6[Vector Memory<br>PostgreSQL + pgvector]
        P7[Embeddings<br>nomic-embed-text via Ollama]
        P8[Traceability<br>CLI + DB logging]
        P9[YAML Pipeline Config<br>Modular control]
        P10[Tooling & CLI<br>ELO visualization]
    end

    subgraph Extensions
        E1[DSPy Integration<br>Composable Signatures & Modules]
        E2[Grafting Logic<br>Merge similar hypotheses]
        E3[Full Pipeline CLI<br>End-to-end execution]
        E4[ELO Visualization<br>Ranking evolution view]
        E5[Local-first Design<br>100% local: Ollama + pgvector]
        E6[Extensible Memory Interface<br>Swap backends]
        E7[Meta Reasoning Hooks<br>Teleprompter & critic ready]
    end

    Start([📄 Paper: AI as a Co-Scientist]) --> Paper_Implementation
    Paper_Implementation --> Extensions
    Extensions --> Future[🔬 Further Research Areas]

    Future --> F1[Teleprompter Optimization]
    Future --> F2[Multi-run Feedback Loops]
    Future --> F3[Experiment Tracking]
    Future --> F4[Evaluation Benchmarks]
    Future --> F5[RAG Integration]
    Future --> F6[Domain-Specific Models]
    Future --> F7[Multi-Agent Dialogues]
    Future --> F8[Exportable Reports]
    Future --> F9[UI / Dashboard]
    Future --> F10[Security / Logging]

    style Start fill:#e1f5fe,stroke:#333
    style Paper_Implementation fill:#fff3e0,stroke:#333
    style Extensions fill:#e8f5e9,stroke:#333
    style Future fill:#fce4ec,stroke:#333

✅ What We’ve Covered from the Paper

(Based on [“AI as a Co-Scientist” — 2502.18864v1])

Appendix 16: Implementing AI as a Co-Scientist: A Research Case Study

Fri, 28 Aug 2026 10:00:00 +0100

Based on:
van der Wal, R., et al. (2024). AI as a Co-Scientist: Hypothesis Generation and Evaluation with LLMs. arXiv:2502.18864v1

    flowchart TD
    Start([📄 Paper: AI as a Co-Scientist]) --> A[📋 Define Stack<br>DSPy, Haystack, pgvector, Ollama, PostgreSQL]
    A --> B[🏗️ Scaffold Project Structure]
    B --> C[🧠 Create BaseAgent Interface]
    C --> D[🔧 Implement Core Agents]
    
    D --> D1[GenerationAgent<br>DSPy GenerateSignature]
    D --> D2[ReflectionAgent<br>DSPy ReflectSignature]
    D --> D3[RankingAgent<br>ELO + JudgeSignature]
    D --> D4[EvolutionAgent<br>EvolveSignature + Grafting]
    D --> D5[MetaReviewAgent<br>MetaReviewSignature]
    
    D1 --> E[🧩 Define Hypothesis Dataclass]
    D2 --> E
    D3 --> E
    D4 --> E
    D5 --> E
    
    E --> F[💾 Implement VectorMemory<br>BaseMemory + pgvector]
    F --> G[🔍 Add WebSearchTool<br>DuckDuckGo search]
    G --> H[⚙️ Build PipelineRunner<br>YAML config-driven]
    H --> I[📊 Add CLI Viewer<br>Ranking trace + ELO evolution]
    I --> J[✅ End-to-End Pipeline]
    J --> K[🧪 Test Suite<br>Pytest for all components]

    style Start fill:#e1f5fe,stroke:#333
    style A fill:#fff3e0,stroke:#333
    style B fill:#e8f5e9,stroke:#333
    style C fill:#fce4ec,stroke:#333
    style D fill:#f3e5f5,stroke:#333
    style D1 fill:#fff9c4,stroke:#333
    style D2 fill:#e0f7fa,stroke:#333
    style D3 fill:#c8e6c9,stroke:#333
    style D4 fill:#ffccbc,stroke:#333
    style D5 fill:#d1c4e9,stroke:#333
    style E fill:#b2dfdb,stroke:#333
    style F fill:#ffcdd2,stroke:#333
    style G fill:#e1f5fe,stroke:#333
    style H fill:#fff3e0,stroke:#333
    style I fill:#e8f5e9,stroke:#333
    style J fill:#c8e6c9,stroke:#333
    style K fill:#4caf50,stroke:#333,color:#fff

You:

Make Experiments Reproducible

Mon, 10 Aug 2026 20:57:00 +0100

By this stage the book contains many systems whose behavior depends on parameters, seeds and implementation choices.

If we cannot reconstruct a run, we cannot really compare it.

Treat configuration as data

from dataclasses import dataclass, asdict


@dataclass(frozen=True)
class ExperimentConfig:
    model: str
    width: int
    height: int
    steps: int
    seed: int
    backend: str
    dtype: str
    params: dict = None

Then extend with model-specific parameters rather than hiding them in notebook cells.

Run Parameter Sweeps and Benchmarks

Mon, 10 Aug 2026 20:58:00 +0100

Once experiments are reproducible, we can stop treating parameters as one-off choices and start mapping behavior systematically.

A parameter sweep is not merely a convenience.

It is a way to turn a model into a measurable landscape.

Build a grid of experiments

Sweep over rules and seeds with owned machinery — Chapter 18’s runner and Chapter 19’s fingerprint, reused exactly:

from itertools import product


def sweep_rules(rules, seeds, width=201, generations=200):
    records = []

    for rule, seed in product(rules, seeds):
        history = run_rule(
            rule,
            width=width,
            generations=generations,
            seed=seed,
            initial="random",
        )
        records.append({
            "rule": rule,
            "seed": seed,
            **fingerprint(history),
        })

    return records

Notice the seed axis.

Capstone — Discover, Measure and Explain a New System

Mon, 10 Aug 2026 21:01:00 +0100

The previous chapter built the laboratory. This one uses it on a single problem.

The book began with one tiny rule applied to one tiny neighborhood.

We end with a different question:

Can we discover a cellular system, characterize its behavior, test its robustness and explain what we actually know about it?

That is the capstone. It uses the ideas rather than listing them: every stage below reuses owned machinery — Lenia configs and kernels (Ch31), discovery scoring (Ch32), outcome classes (Ch33), the measurement stack (Ch18–22), perturbation discipline (Ch35), search strategy (Ch24–25), and experiment records (Ch55).

What Does a Preference Know About the Future?

Fri, 10 Jul 2026 11:32:58 +0100

We Trained a Model on Editorial Choices to See Whether It Learned What Happened Next

Most preference-learning systems use a choice to change the future.

A model produces two responses. A human selects one. The chosen response becomes positive evidence, the rejected response becomes negative evidence, and training makes outputs resembling the chosen response more likely.

The preference acts as an instruction:

Produce more things like this.

I wanted to know whether the same choice could also function as evidence.

The Preference Was Only the Beginning

Tue, 07 Jul 2026 00:00:00 +0100

A preference is not only a label on what just happened. When the decision belongs to a continuing trajectory, it can also be evidence about what happens next.

Abstract

Most preference-learning systems stop at the choice.

A model produces two responses. A human selects one. The chosen response becomes positive evidence, the rejected response becomes negative evidence, and the training system moves on.

The work itself usually continues.

The selected answer may later be revised, partially retained, contradicted or abandoned. The rejected alternative may reveal a constraint that remains active long after the immediate decision. The preference is therefore not necessarily the outcome. It may be an event inside a longer trajectory.

The Asset‑Price State: How the U.S. Fiscal Machine Now Depends on Rising Markets

Wed, 17 Jun 2026 17:18:52 +0100

What this post argues

The U.S. does not merely like a high stock market. It increasingly needs one.

The stock market has become an amplifier inside the largest federal revenue pipe: individual income tax. When asset prices rise, capital gains, stock compensation, options, bonuses, business equity, and other asset-sensitive income strengthen federal receipts. When the market falls hard, that same amplifier runs in reverse.

This matters because the real debt problem is not debt alone. It is interest cost relative to federal revenue. If interest grows faster than normal revenue, the system becomes more dependent on asset inflation, AI valuations, national champions, and foreign capital to keep the fiscal machine stable.

Warranted Search: When AI Must Prove Before It Looks

Tue, 02 Jun 2026 18:54:26 +0100

TL;DR

Modern AI systems often retrieve nearby text and generate confident answers, but that is not the same as proof. A citation can be real, the answer can be correct, and the evidence can still fail to support the claim.

This post argues for warranted search: a scoped, claim-driven form of intelligent grep. Instead of asking an AI to rummage through a corpus, we give it a specific claim, a bounded search warrant, and a limited set of safe operations.

AI and the End of Easy Growth

Tue, 19 May 2026 09:52:38 +0100

Constraint, Continuity, and Human Agency

1. Introduction: The Strange Feeling Around AI

Artificial Intelligence is clearly real. The infrastructure buildout is enormous, adoption is accelerating, and AI systems are already reshaping software, research, administration, and knowledge work. Yet for many people, daily life feels strangely unchanged. Housing remains expensive, wages remain under pressure, debt continues to rise, and institutions across the West increasingly appear constrained rather than confident.

This post explores the possibility that AI is currently being deployed less as a technology of broad civilizational expansion and more as a technology of continuity management, a way for highly complex systems under demographic, economic, and energetic pressure to optimize themselves, compress costs, and maintain stability in an era where traditional growth models are becoming harder to sustain.

Canada: When Interest Meets Reliable Revenue

Wed, 15 Apr 2026 12:55:18 +0100

Executive Summary

Canada’s fiscal position looks stable on paper. Headline interest costs consume only ~10.6% of federal revenue. But this ratio masks a structural reality: the engine that drove revenue growth has stalled, and the cost of past debt is rising faster than the system can generate new fiscal space.

For decades, population expansion concealed weak per-capita productivity. In 2025, that demographic engine stopped. At the same time, Canada does not fully capture or retain the economic value it produces, due to commodity pricing discounts, single-customer trade concentration, and high-skill outflows. When these factors are applied to the revenue base, the effective denominator shrinks.

From Fuel Protests to Fiscal Risk: What’s Really Happening in Ireland

Tue, 14 Apr 2026 17:53:37 +0100

Executive Summary

This post applies a simple, testable framework to Ireland’s fiscal system:

Fiscal constraint emerges when the cost of debt rises relative to the revenue supporting it.

In large, stable systems like the United States, this dynamic unfolds gradually. Ireland presents a different case.

While headline metrics suggest strength, three structural factors create a distinct risk profile:

Revenue composition: A significant portion derives from multinational activity and is not fully under domestic control.
Measurement distortion: The effective economic base (GNI*) is ~43% smaller than GDP implies.
Debt repricing: Existing debt is being refinanced at materially higher interest rates.

These factors introduce a critical refinement to the model:

✨ TINY CRITICS: Lightweight Reasoning Checks for Large AI Systems

Sat, 29 Nov 2025 00:06:42 +0000

🥹 0. TL;DR

Large language models write fluent explanations even when they’re wrong. Verifying their reasoning usually requires another LLM slow, expensive, and circular.

We needed something different:

A miniature reasoning critic <50 KB trained on synthetic reasoning mistakes, able to instantly detect broken reasoning in much larger models.

The Tiny Critic:

trains on GSM8K-style reasoning traces generated by DeepSeek or Mistral
uses FrontierLens, and Visual Policy Maps (VPMs) to convert reasoning into canonical numerical features
is just a logistic regression with ~30 parameters
runs in microseconds
plugs into any agent
dramatically improves InitAgent, R1-Loops, and research-planning stability

This post tells the full story how we built it, why it works, and what we learned about the shape of reasoning.

The Space Between Models Has Holes: Mapping the AI Gap

Wed, 22 Oct 2025 20:30:36 +0100

🌌 Summary

What if the most valuable insights in AI evaluation aren’t in model agreements, but in systematic disagreements?

This post reveals that the “gap” between large and small reasoning models contains structured, measurable intelligence about how different architectures reason. We demonstrate how to transform model disagreements from a problem into a solution, using the space between models to make tiny networks behave more like their heavyweight counterparts.

We start by assembling a high-quality corpus (10k–50k conversation turns), score it with a local LLM to create targets, and train both HRM and Tiny models under identical conditions. Then we run fresh documents through both models, collecting not just final scores but rich auxiliary signals (uncertainty, consistency, OOD detection, etc.) and visualize what these signals reveal.

🔦 Phōs: Visualizing How AI Learns and How to Build It Yourself

Thu, 09 Oct 2025 00:30:36 +0100

“The eye sees only what the mind is prepared to comprehend.” Henri Bergson

🔍 We Finally See Learning

For decades, we’ve measured artificial intelligence with numbers loss curves, accuracy scores, reward signals.
We’ve plotted progress, tuned hyperparameters, celebrated benchmarks.

But we’ve never actually seen learning happen.

Not really.

Sure, we’ve visualized attention maps or gradient flows but those are snapshots, proxies, not processes.

What if we could watch understanding emerge not as a number going up, but as a pattern stabilizing across time?
What if reasoning itself left a visible trace?

Detecting AI-Generated Text: Challenges and Solutions

Thu, 13 Mar 2025 13:28:24 +0000

Summary

Artificial Intelligence (AI) has revolutionized the way we generate and consume text. From chatbots crafting customer responses to AI-authored articles, artificial intelligence is reshaping how we create and consume content. As AI-generated text becomes indistinguishable from human writing, distinguishing between the two has never been more critical. Here are some of the reasons it is important to be able to verify the source of information:

Preventing plagiarism
Maintaining academic integrity
Ensuring transparency in content creation
If AI models are repeatedly trained on AI-generated text, their quality may degrade over time.

In this blog post, we’ll explore the current most effective methods for detecting AI-generated text.

Shakespeare and the Bible: An AI Investigation

Tue, 11 Mar 2025 22:47:36 +0000

Summary

Could the greatest playwright of all time have secretly shaped one of the most influential religious texts in history? Some believe William Shakespeare left his mark on the King James Bible hidden in plain sight. With the power of AI, we’ll investigate whether there’s any truth to this conspiracy.

You can read about the conspiracy here: