# shtosti/accelerated-drift architecture

> Architecture diagram of the GitHub repository shtosti/accelerated-drift, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-05). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/shtosti/accelerated-drift
- Repository: https://github.com/shtosti/accelerated-drift

## Overview

This is a research pipeline for measuring stylistic changes in arXiv and medRxiv preprints around LLM adoption. Researchers invoke configured collection, preprocessing, analysis, and visualization commands; the result is statistical tables and publication figures, with topic-level and cross-topic comparisons as additional analysis branches. The graph follows documented stages and source-index references. Some internal implementation and external API wiring was not sampled, so uncertain links are omitted.

Main entry points: [`src/not_an_llm/pipelines/analyze.py`](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/pipelines/analyze.py) (Corpus analysis), [`src/not_an_llm/cli.py`](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/cli.py) (Command interface), [`src/not_an_llm/pipelines/collect.py`](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/pipelines/collect.py) (Record collection), [`src/not_an_llm/pipelines/visualize.py`](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/pipelines/visualize.py) (Figure generation)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_control["Runtime control"]
  node_cli["Command interface<br/>[cli.py]"]
  node_config["Run configuration<br/>[config.py]"]
end

subgraph group_ingest["Corpus preparation"]
  node_collect["Record collection<br/>[collect.py]"]
  node_arxiv["arXiv client<br/>[arxiv.py]"]
  node_medarxiv["medRxiv client<br/>[medarxiv.py]"]
  node_biorxiv["bioRxiv client<br/>[biorxiv.py]"]
  node_semantic["Semantic Scholar client"]
  node_raw[("Paper records")]
  node_preprocess["Text preprocessing<br/>[preprocess.py]"]
  node_text["Text normalization<br/>[text.py]"]
  node_processed[("Prepared corpus")]
end

subgraph group_analysis["Statistical analysis"]
  node_analyze["Corpus analysis<br/>[analyze.py]"]
  node_extract["Feature extraction"]
  node_readability["Readability metrics<br/>[readability.py]"]
  node_groups["Feature families<br/>[feature_groups.py]"]
  node_trends["Trend aggregation<br/>[trends.py]"]
  node_its["Interrupted time series"]
  node_topics["Topic modeling<br/>[model.py]"]
  node_topic_pipeline["Topic reporting<br/>[pipeline.py]"]
  node_comparison["Topic comparison<br/>[comparison.py]"]
end

subgraph group_outputs["Reports and outputs"]
  node_analysis_data[("Analysis tables")]
  node_visualize["Figure generation<br/>[visualize.py]"]
  node_visuals[("Rendered figures")]
end

node_researcher(("Researcher"))
node_preprint_apis{{"Preprint APIs"}}
node_embedding_models{{"Embedding models"}}

node_researcher -->|"runs commands"| node_cli
node_cli -->|"loads settings"| node_config
node_cli -->|"dispatches collection"| node_collect
node_cli -->|"dispatches preprocessing"| node_preprocess
node_cli -->|"dispatches analysis"| node_analyze
node_cli -->|"dispatches visualization"| node_visualize
node_cli -->|"dispatches comparison"| node_comparison
node_collect -->|"reads settings"| node_config
node_collect -->|"queries"| node_arxiv
node_collect -->|"queries"| node_medarxiv
node_collect -->|"queries"| node_biorxiv
node_collect -->|"queries"| node_semantic
node_arxiv -.->|"requests records"| node_preprint_apis
node_medarxiv -.->|"requests records"| node_preprint_apis
node_collect -->|"writes records"| node_raw
node_preprocess -->|"reads settings"| node_config
node_preprocess -->|"normalizes text"| node_text
node_preprocess -->|"reads records"| node_raw
node_preprocess -->|"writes corpus"| node_processed
node_analyze -->|"reads settings"| node_config
node_analyze -->|"reads corpus"| node_processed
node_analyze -->|"extracts features"| node_extract
node_analyze -->|"computes metrics"| node_readability
node_analyze -->|"uses feature families"| node_groups
node_analyze -->|"aggregates trends"| node_trends
node_analyze -->|"fits models"| node_its
node_analyze -->|"runs topic analysis"| node_topic_pipeline
node_topic_pipeline -->|"assigns topics"| node_topics
node_topics -.->|"encodes documents"| node_embedding_models
node_analyze -->|"writes statistics"| node_analysis_data
node_topic_pipeline -->|"writes topic reports"| node_analysis_data
node_comparison -->|"computes ITS"| node_its
node_comparison -->|"reads comparison inputs"| node_analysis_data
node_visualize -->|"reads settings"| node_config
node_visualize -->|"reads tables"| node_analysis_data
node_visualize -->|"plots trends"| node_trends
node_visualize -->|"plots ITS results"| node_its
node_visualize -->|"writes figures"| node_visuals

click node_cli "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/cli.py"
click node_config "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/config.py"
click node_collect "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/pipelines/collect.py"
click node_arxiv "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/clients/arxiv.py"
click node_medarxiv "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/clients/medarxiv.py"
click node_biorxiv "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/clients/biorxiv.py"
click node_semantic "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/clients/semantic_scholar.py"
click node_raw "https://github.com/shtosti/accelerated-drift/tree/main/data/raw"
click node_preprocess "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/pipelines/preprocess.py"
click node_text "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/preprocessing/text.py"
click node_processed "https://github.com/shtosti/accelerated-drift/tree/main/data/processed"
click node_analyze "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/pipelines/analyze.py"
click node_extract "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/analysis/feature_extractor.py"
click node_readability "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/analysis/readability.py"
click node_groups "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/analysis/feature_groups.py"
click node_trends "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/analysis/trends.py"
click node_its "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/analysis/interrupted_time_series.py"
click node_topics "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/analysis/topic_modeling/model.py"
click node_topic_pipeline "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/analysis/topic_modeling/pipeline.py"
click node_comparison "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/analysis/topic_modeling/comparison.py"
click node_analysis_data "https://github.com/shtosti/accelerated-drift/tree/main/data/analysis"
click node_visualize "https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/pipelines/visualize.py"
click node_visuals "https://github.com/shtosti/accelerated-drift/tree/main/data/visuals"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_cli,node_config toneBlue
class node_collect,node_arxiv,node_medarxiv,node_biorxiv,node_semantic,node_raw,node_preprocess,node_text,node_processed toneAmber
class node_analyze,node_extract,node_readability,node_groups,node_trends,node_its,node_topics,node_topic_pipeline,node_comparison,node_preprint_apis toneMint
class node_analysis_data,node_visualize,node_visuals toneRose
class node_researcher,node_embedding_models toneIndigo
```

## Components

### Runtime control

- [Command interface](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/cli.py): `src/not_an_llm/cli.py`
- [Run configuration](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/config.py): `src/not_an_llm/config.py`

### Corpus preparation

- [Record collection](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/pipelines/collect.py): `src/not_an_llm/pipelines/collect.py`
- [arXiv client](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/clients/arxiv.py): `src/not_an_llm/clients/arxiv.py`
- [medRxiv client](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/clients/medarxiv.py): `src/not_an_llm/clients/medarxiv.py`
- [bioRxiv client](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/clients/biorxiv.py): `src/not_an_llm/clients/biorxiv.py`
- [Semantic Scholar client](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/clients/semantic_scholar.py): `src/not_an_llm/clients/semantic_scholar.py`
- [Paper records](https://github.com/shtosti/accelerated-drift/tree/main/data/raw): `data/raw`
- [Text preprocessing](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/pipelines/preprocess.py): `src/not_an_llm/pipelines/preprocess.py`
- [Text normalization](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/preprocessing/text.py): `src/not_an_llm/preprocessing/text.py`
- [Prepared corpus](https://github.com/shtosti/accelerated-drift/tree/main/data/processed): `data/processed`

### Statistical analysis

- [Corpus analysis](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/pipelines/analyze.py): `src/not_an_llm/pipelines/analyze.py`
- [Feature extraction](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/analysis/feature_extractor.py): `src/not_an_llm/analysis/feature_extractor.py`
- [Readability metrics](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/analysis/readability.py): `src/not_an_llm/analysis/readability.py`
- [Feature families](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/analysis/feature_groups.py): `src/not_an_llm/analysis/feature_groups.py`
- [Trend aggregation](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/analysis/trends.py): `src/not_an_llm/analysis/trends.py`
- [Interrupted time series](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/analysis/interrupted_time_series.py): `src/not_an_llm/analysis/interrupted_time_series.py`
- [Topic modeling](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/analysis/topic_modeling/model.py): `src/not_an_llm/analysis/topic_modeling/model.py`
- [Topic reporting](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/analysis/topic_modeling/pipeline.py): `src/not_an_llm/analysis/topic_modeling/pipeline.py`
- [Topic comparison](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/analysis/topic_modeling/comparison.py): `src/not_an_llm/analysis/topic_modeling/comparison.py`

### Reports and outputs

- [Analysis tables](https://github.com/shtosti/accelerated-drift/tree/main/data/analysis): `data/analysis`
- [Figure generation](https://github.com/shtosti/accelerated-drift/blob/main/src/not_an_llm/pipelines/visualize.py): `src/not_an_llm/pipelines/visualize.py`
- [Rendered figures](https://github.com/shtosti/accelerated-drift/tree/main/data/visuals): `data/visuals`

### Other components

- **Researcher**
- **Preprint APIs**
- **Embedding models**

## Connections

- Researcher → Command interface: runs commands (evidence: `README.md`)
- Command interface → Run configuration: loads settings (evidence: `src/not_an_llm/cli.py`)
- Command interface → Record collection: dispatches collection (evidence: `src/not_an_llm/cli.py`)
- Command interface → Text preprocessing: dispatches preprocessing (evidence: `src/not_an_llm/cli.py`)
- Command interface → Corpus analysis: dispatches analysis (evidence: `src/not_an_llm/cli.py`)
- Command interface → Figure generation: dispatches visualization (evidence: `src/not_an_llm/cli.py`)
- Command interface → Topic comparison: dispatches comparison (evidence: `src/not_an_llm/cli.py`)
- Record collection → Run configuration: reads settings (evidence: `src/not_an_llm/pipelines/collect.py`)
- Record collection → arXiv client: queries (evidence: `src/not_an_llm/pipelines/collect.py`)
- Record collection → medRxiv client: queries (evidence: `src/not_an_llm/pipelines/collect.py`)
- Record collection → bioRxiv client: queries (evidence: `src/not_an_llm/pipelines/collect.py`)
- Record collection → Semantic Scholar client: queries (evidence: `src/not_an_llm/pipelines/collect.py`)
- arXiv client → Preprint APIs: requests records
- medRxiv client → Preprint APIs: requests records
- Record collection → Paper records: writes records (evidence: `README.md`)
- Text preprocessing → Run configuration: reads settings (evidence: `src/not_an_llm/pipelines/preprocess.py`)
- Text preprocessing → Text normalization: normalizes text (evidence: `src/not_an_llm/pipelines/preprocess.py`)
- Text preprocessing → Paper records: reads records (evidence: `README.md`)
- Text preprocessing → Prepared corpus: writes corpus (evidence: `README.md`)
- Corpus analysis → Run configuration: reads settings (evidence: `src/not_an_llm/pipelines/analyze.py`)
- Corpus analysis → Prepared corpus: reads corpus (evidence: `README.md`)
- Corpus analysis → Feature extraction: extracts features (evidence: `src/not_an_llm/pipelines/analyze.py`)
- Corpus analysis → Readability metrics: computes metrics (evidence: `src/not_an_llm/pipelines/analyze.py`)
- Corpus analysis → Feature families: uses feature families (evidence: `src/not_an_llm/pipelines/analyze.py`)
- Corpus analysis → Trend aggregation: aggregates trends (evidence: `src/not_an_llm/pipelines/analyze.py`)
- Corpus analysis → Interrupted time series: fits models (evidence: `README.md`)
- Corpus analysis → Topic reporting: runs topic analysis (evidence: `src/not_an_llm/pipelines/analyze.py`)
- Topic reporting → Topic modeling: assigns topics (evidence: `src/not_an_llm/analysis/topic_modeling/pipeline.py`)
- Topic modeling → Embedding models: encodes documents (evidence: `README.md`)
- Corpus analysis → Analysis tables: writes statistics (evidence: `README.md`)
- Topic reporting → Analysis tables: writes topic reports (evidence: `src/not_an_llm/analysis/topic_modeling/pipeline.py`)
- Topic comparison → Interrupted time series: computes ITS (evidence: `src/not_an_llm/analysis/topic_modeling/comparison.py`)
- Topic comparison → Analysis tables: reads comparison inputs
- Figure generation → Run configuration: reads settings (evidence: `src/not_an_llm/pipelines/visualize.py`)
- Figure generation → Analysis tables: reads tables (evidence: `README.md`)
- Figure generation → Trend aggregation: plots trends (evidence: `src/not_an_llm/pipelines/visualize.py`)
- Figure generation → Interrupted time series: plots ITS results (evidence: `src/not_an_llm/pipelines/visualize.py`)
- Figure generation → Rendered figures: writes figures (evidence: `README.md`)
