eSapiens's DEREK Module: Deep Extraction & Reasoning Engine for Knowledge with LLMs

TL;DR

The paper introduces the DEREK module, a secure, scalable retrieval-augmented generation system for enterprise document question answering, combining advanced retrieval, reranking, and verification to ensure accurate and traceable answers.

Contribution

It presents a novel, end-to-end pipeline integrating heterogeneous content ingestion, hybrid retrieval, reranking, and a verifier for enterprise-grade document QA with minimal operational overhead.

Findings

Improved Recall@50 by ~1 percentage point on LegalBench

Boosted Precision@10 by ~7 percentage points with hybrid+rerank

Verifier increased TRACe Utilization above 0.50 and reduced unsupported statements below 3%

Abstract

We present the DEREK (Deep Extraction & Reasoning Engine for Knowledge) Module, a secure and scalable Retrieval-Augmented Generation pipeline designed specifically for enterprise document question answering. Designed and implemented by eSapiens, the system ingests heterogeneous content (PDF, Office, web), splits it into 1,000-token overlapping chunks, and indexes them in a hybrid HNSW+BM25 store. User queries are refined by GPT-4o, retrieved via combined vector+BM25 search, reranked with Cohere, and answered by an LLM using CO-STAR prompt engineering. A LangGraph verifier enforces citation overlap, regenerating answers until every claim is grounded. On four LegalBench subsets, 1000-token chunks improve Recall@50 by approximately 1 pp and hybrid+rerank boosts Precision@10 by approximately 7 pp; the verifier raises TRACe Utilization above 0.50 and limits unsupported statements to less than 3%. All components run in containers, enforce end-to-end TLS 1.3 and AES-256. These results demonstrate that the DEREK module delivers accurate, traceable, and production-ready document QA with minimal operational overhead. The module is designed to meet enterprise demands for secure, auditable, and context-faithful retrieval, providing a reliable baseline for high-stakes domains such as legal and finance.