pprag 0.1.0

Proxy-Pointer RAG suite for text, multimodal, and cross-document comparison workflows

Proxy-Pointer Suite -- Text, Multimodal RAG, and Cross-Document Comparison 🔍

Structural RAG for Complex Documents — A high-fidelity retrieval pipeline that uses document hierarchy as the primary retrieval anchor, eliminating "hallucination by chunking." Proxy-Pointer indexes structural pointers (breadcrumbs like Paper > Section > Sub-section) rather than raw text fragments, ensuring the LLM always understands exactly where it is in a document.

Retrieve precise text, get grounded visual citations, or perform Agentic section-by-section document comparisons.

---

Three Implementations, One Architecture

Feature	Text-Only	MultiModal	DocComparator
Core Goal	Maximum precision for text-based RAG	Unified reasoning across text & visuals	Agentic Cross-Document Comparison
Input	Structured Markdown (LlamaParse)	Markdown + Figures/Tables (Adobe Extract)	PDF or MD (Mixed format supported)
Output	Text-based answers	Text +$\color{#15803d}{\textsf{\textbf{AI-Verified Visual Evidence}}}$ 🖼️	Side-by-side analytical reports
LLM	Gemini 3.1 Flash-Lite	Gemini 3.1 Flash-Lite	Gemini 3 Flash
Embeddings	gemini-embedding-001 (1536d)	gemini-embedding-001 (1536d)	gemini-embedding-001 (1536d)
Vision	—	✅ Gemini 3.1 Flash-Lite	—
Retrieval	Structural re-ranking (k=5)	Anchor-aware re-ranking + image selection	Multi-Stage Proxy-Pointer retrieval
Benchmark	100% on FinanceBench	96% across 20-query, 5-paper suite	N/A (Dynamic Agentic Evaluation)
Use Case	10-K Financials, Legal, Documentation	Anything with Images, Diagrams, Charts	Credit Agreements, Contracts, Research Papers
Interface	CLI / Python API	Streamlit UI with visual citations	Streamlit UI with markdown export

---

How It Works

graph TD
    A[Documents] -->|PDF Extraction| B[Markdown]
    B -->|Tree Builder| C[Structure Trees]
    C -->|Noise Filter| D[Clean Nodes]
    D -->|Embed + Index| E[FAISS]
    E -->|"Query, Dedup, Re-Rank"| F[Top Sections]
    F -->|Synthesize + Cite| G[Grounded Answer]

1. Structure trees map every section, sub-section, figure, and table in a document
2. Noise filtering removes TOC, glossaries, and boilerplate using an LLM
3. Broad vector recall (k=200) retrieves candidates, then LLM re-ranking selects the best structural matches
4. Full section loading gives the synthesizer complete context — not truncated chunks
5. (MultiModal only) Anchor-aware retrieval surfaces figures/tables physically linked to retrieved sections

---

Which One Should I Use?

Text-Only — Best when your documents are purely text-based and the hierarchy (e.g., Signatory > Item 1A > Risk Factors) is the only context needed. Proven at 100% accuracy on financial 10-K filings.

MultiModal — Best when your documents contain diagrams, charts, and tables that are essential to the answer. Uses anchor-aware retrieval to surface the exact images tied to a technical discussion, tested across 5 research papers (CLIP, GaLore, NemoBot, VectorFusion, VectorPainter).

DocComparator — Best when you need to perform deep, section-by-section comparisons between two complex documents. Uses Agentic RAG and targeted personas (like Senior Legal Counsel) to untangle legal trade-offs and methodological differences beyond surface-level keyword matching.

---

Architecture Deep Dive

For the full technical story behind the architecture:

1. Proxy-Pointer Framework for Structure-Aware Enterprise Document Intelligence — Hierarchical understanding and comparison of contracts, research papers, and more
2. Proxy-Pointer RAG: Multimodal Answers Without Multimodal Embeddings — Structure is all you need
3. Proxy-Pointer RAG: Structure Meets Scale — 100% Accuracy with Smarter Retrieval — Scaling to multi-document, LLM re-ranking, and benchmark results
4. Proxy-Pointer RAG: Achieving Vectorless Accuracy at Vector RAG Scale and Cost — Core architecture & the pointer-based retrieval idea

---

Quick Start

For Users (Via PyPI)

We recommend creating a virtual environment first. Install only the modality you need:

python -m venv venv
# Windows: .\venv\Scripts\activate | Mac/Linux: source venv/bin/activate

pip install pprag                 # minimal CLI shell
pip install "pprag[text]"         # text-only structural RAG
pip install "pprag[multimodal]"   # multimodal RAG with visual citations
pip install "pprag[compare]"      # cross-document comparison
pip install "pprag[full]"         # all modalities

Then choose the workflow from the pprag CLI:

pprag text index --fresh
pprag text ask

pprag multimodal index --fresh
pprag multimodal serve

pprag compare serve

The default install intentionally stays lightweight. If you run a modality without its optional dependencies, the CLI prints the exact extra to install, for example pip install "pprag[multimodal]".

For Developers (Local Repository)

If you want to tinker with the code, view the raw prompts, or use the sample data, we recommend cloning the repository. This project uses uv for lightning-fast dependency management (you can install it via pip install uv).

git clone https://github.com/partha-sarkar/proxy-pointer.git
cd proxy-pointer

# Sync the isolated environment with all dependencies (uv creates the venv automatically)
uv sync --all-extras

# Prefix commands with `uv run` to execute them inside the automatic environment
uv run pprag text index --fresh
uv run pprag text ask

Each implementation also has its own self-contained README with a 5-minute quickstart:

• Text-Only → Get Started
• MultiModal → Get Started
• DocComparator → Get Started

All include sample data so you can clone, build the index, an start exploring immediately.

---

Author

Partha Sarkar

Contact

• GitHub Issues: For bug reports
• General Questions: Reach out on LinkedIn or Email

---

License

© 2026 Partha Sarkar (Proxy-Pointer). Licensed under MIT.