Pollagraph

Knowledge Acquisition for the Language-Model Era

Language models have the text. Pollagraph adds human reasoning.

Writing in Parallel gives multiple people the same carefully designed prompt and captures their independent reasoning in structured paragraphs.

One Prompt. Multiple Minds. Structured Human Reasoning.

Why Pollagraph

Language Models Already Have Enormous Quantities of Text

Pollagraph adds something different: human reasoning deliberately produced around known questions under known conditions.

Existing text gives language models extraordinary breadth. Pollagraph creates complementary reasoning data by asking people to respond independently to carefully designed questions.

Existing text captures what people have written. Pollagraph asks people to reason.

The Principle

Deliberately Acquire Human Reasoning for AI

Classic expert systems established an important idea: human knowledge could be deliberately acquired for computers.

Knowledge engineers interviewed experts, captured what they knew, and organized that knowledge for machine use.

Pollagraph carries that principle forward.

Instead of interviewing one expert, Pollagraph asks multiple people the same carefully designed question and preserves their independent reasoning together.

The Method

The Prompt Becomes the Interview

Writing in Parallel is Pollagraph's method for deliberately acquiring human reasoning.

Multiple people receive the same carefully designed prompt. Each person responds independently in a structured paragraph.

Because everyone begins with the same question, the responses share a common point of origin.

Because everyone reasons independently, different perspectives, assumptions, judgments, and reasoning paths emerge.

One prompt. Multiple minds. Multiple reasoning paths.

Participate in a Writing in Parallel Session

Writing in Parallel

One common question. Independent human reasoning.

Writer 1 follows one reasoning path.
Writer 2 sees the question differently.
Writer 3 introduces another perspective.
The common prompt keeps the responses connected.

The Relationship

Alignment Makes the Responses Comparable. Variation Makes Them Valuable.

Pollagraph calls the relationship among these independent responses Aligned Variation.

Alignment

Every participant responds to the same question. The responses therefore share a known point of origin.

Variation

Every participant reasons independently, bringing different judgment, experience, assumptions, and interpretation to the question.

Aligned Variation

Different reasoning paths remain connected through the common prompt that produced them.

The differences are not noise. They are part of the knowledge.

The Product

From One Question to a Structured Reasoning Dataset

A standard Pollagraph reasoning set makes the idea concrete.

1

Common Prompt

One carefully designed question establishes the common point of origin.

10

Independent Writers

Ten people consider the same question independently.

10

Structured Paragraphs

Each writer contributes one structured response.

10

Reasoning Paths

Different approaches to the same question remain visible.

1

Aligned Reasoning Set

The responses remain connected through their common prompt.

Pollagraph does not merely collect ten paragraphs. It preserves ten independent responses to the same question as one related reasoning set.

The Data

Structured Human Reasoning Data

Pollagraph calls the resulting data Structured Human Reasoning Data, or SHRD.

SHRD preserves the reasoning and its context.

The value comes not only from what each person writes, but from the known relationship among the responses.

The prompt is known
The response structure is known
The responses share a common question
The responses are independently produced
The reasoning paths remain connected
The production conditions can be documented
SHRD preserves Aligned Variation as structured data.

From Expert Systems to Language Models

Knowledge Acquisition Is the Bridge

Pollagraph does not recreate the architecture of the classic expert system. It carries forward the principle of deliberately acquiring human knowledge for machines.

Classic Expert System

Expert
↓
Interview
↓
Knowledge Acquisition
↓
Knowledge Base
↓
Expert System

Pollagraph

Multiple Human Reasoners
↓
Common Prompt
↓
Writing in Parallel
↓
Aligned Variation
↓
SHRD
↓
Language-Model Development
The technology has changed.
The principle of deliberate knowledge acquisition has not.

The Pollagraph Thesis

Complement Existing Text With Deliberately Acquired Human Reasoning

Language models already have enormous quantities of human language.

Pollagraph proposes something complementary: human reasoning deliberately produced around known questions under known conditions.

Writing in Parallel creates the responses. Aligned Variation describes their relationship. SHRD preserves that relationship as structured data.

Can deliberately acquired human reasoning complement the enormous quantities of existing text used by language models?

Pollagraph exists to pursue that possibility.

Where the Data Begins

Writing in Parallel Sessions

A Writing in Parallel session puts Pollagraph's knowledge-acquisition method into practice.

Participants receive common prompts and respond independently in structured paragraphs.

The common prompt creates alignment. The participants create variation. Writing in Parallel captures both.

The resulting responses can be organized into SHRD datasets around particular questions, subjects, problems, or knowledge domains.

The Pollagraph Process

From Question to Dataset

Ask a common question.
Capture independent reasoning.
Preserve Aligned Variation.
Build Structured Human Reasoning Data.

The Pollagraph System

Four Ideas. One System.

Each part of Pollagraph has a distinct role.

Principle

Knowledge Acquisition

Deliberately acquire human knowledge for machines.

Method

Writing in Parallel

Ask multiple people the same carefully designed question.

Relationship

Aligned Variation

Preserve different reasoning paths around a common point of origin.

Data

SHRD

Structure the responses and preserve their relationships.

Knowledge Acquisition → Writing in Parallel → Aligned Variation → Structured Human Reasoning Data

Pollagraph

Build the Knowledge AI Needs

Pollagraph connects institutions that can produce structured human reasoning with AI organizations interested in exploring that reasoning as a complementary source of data.

One Prompt.
Multiple Minds.
Aligned Variation.
Structured Human Reasoning.

For Institutions

Host a Writing in Parallel Session

Bring participants together around common prompts and participate in producing structured human reasoning data.

Host a Session

For AI Organizations

Explore SHRD Datasets

Explore deliberately produced human reasoning as a complementary source of data for language-model development.

Explore SHRD