Dual-layer knowledge graph connects fragmented pharmaceutical process-development documents

The system combines document-level retrieval with ontology-based links, achieving 95% multiple-choice accuracy and an 85% LLM-judge pass rate on proprietary reports.

PaperIndustrycs.AIarXiv:2609.11493v1
Reza Amirmoshiri · Faryad Sahneh · Yasser Jangjou

Sanofi US

Research Digest··2 min read
Amirmoshiri, Sahneh and Jangjou built an agentic platform that transforms heterogeneous Chemistry, Manufacturing and Controls documents into a queryable, provenance-linked knowledge graph. On 505 questions drawn from 38 reports for one Sanofi small-molecule program, the lexical retrieval layer performed strongly on direct questions but weakened on comparative and corpus-wide queries.

What they did

The authors created a dual-layer knowledge graph for pharmaceutical process development. Its base layer preserves documents in a Document–Section–Chunk hierarchy, including digital, scanned, handwritten and multilingual material; its intelligence layer extracts ontology-aligned entities and connects concepts across documents and projects while retaining links to source passages.

LLM agents route questions to the appropriate retrieval path, from lexical lookup to domain-graph reasoning. The authors evaluated the lexical layer using a three-tier protocol and 505 questions curated from 38 proprietary development reports covering a Sanofi small-molecule program.

Key findings

  • Tier-1 multiple-choice accuracy reached 95%, indicating reliable retrieval and answer selection on the curated test set.
  • Under the stricter Tier-2 evaluation by an LLM judge, the pass rate fell to 85%.
  • Comparative and corpus-wide questions produced more failures than simpler lookups, exposing weaknesses that multiple-choice accuracy did not capture.
  • A router agent selected between the lexical and ontology-based graph layers according to question type.

Why it matters

CMC knowledge is commonly dispersed across incompatible documents, making technology transfer, investigation and regulatory preparation labor-intensive. The architecture offers a way to preserve source-level traceability while also supporting structured queries across documents, and the evaluation protocol provides a practical template for testing retrieval systems when the underlying industrial corpus cannot be released publicly.

Caveats

The reported benchmark covers 38 reports from a single proprietary small-molecule program, so performance may not generalize to other projects, organizations or document distributions. The quantitative results primarily evaluate the lexical layer rather than the full domain graph and routing system, and the Tier-2 score depends on an LLM judge whose calibration and reproducibility are not described in the supplied material.

§

Analysis

The work combines two common approaches to enterprise retrieval: chunk-level retrieval that preserves document context and ontology-backed graphs that normalize concepts across sources. Its most useful contribution may be the separation of direct retrieval reliability from harder comparative and corpus-level reasoning, where aggregate multiple-choice scores can conceal important failure modes.

newspaper

Research Digest

Articles published under the Zotpaper byline are synthesized from multiple source publications by our AI editor and reviewed by our editorial process. Each story combines reporting from credible outlets to give readers a balanced, comprehensive view.