Summary
Scientific papers on materials like superconductors are published faster than people can read them, and usual search tools rely on simple word matching that misses important details. The authors created a system that reads these papers, pulls out specific data about materials and their properties, and helps answer questions more accurately by using special knowledge about the field. They tested their extraction method against simpler techniques and showed it works much better at finding correct material-property pairs. However, adding the extracted data to the search didn’t always improve results due to noisy information, showing how hard it is to find the exact supporting evidence.
What this means in practice
- •For materials informatics teams: Search and analyze superconductivity publications with extracted structured data for more precise retrieval of materials and properties.
- •For scientific literature search providers: Enhance search engines with domain-aware extraction and hybrid retrieval techniques specialized for physics PDFs to improve answer accuracy.$Commercial implications: This paper enables new search tools selling tailored physics literature access with richer, structured content and better question answering.
Authors
Ivan Iorsh, Aleksandr Kapitonov, Ivan Filianin, Matvey Lesonen, Nikita Chizhov
Abstract
Scientific literature in materials physics is growing faster than researchers can manually inspect it, yet most retrieval tools rely on keyword matching or generic question answering that ignores domain-specific structure in PDF files. This paper presents an end-to-end pipeline that transforms arXiv superconductivity publications into a searchable corpus with structured metadata and retrieval-augmented question answering. The system combines thematic corpus construction, domain-aware extraction of materials and superconducting transition temperatures, and hybrid retrieval that merges dense chunk embeddings with structured publication fields. We test whether normalized domain attributes improve evidence retrieval and answer groundedness relative to lexical-only and dense text-only methods. Extraction is evaluated on a manually verified gold set of 59 superconducting materials, 46 of which have confirmed experimental critical temperatures. On the 33-material primary positive subset, domain-aware pair accuracy reaches 45.5% versus 9.1% for a regex baseline. Retrieval and generation are compared under a fixed language model (qwen2.5:7b-instruct) on a 100-question benchmark, varying only the retrieval mechanism: BM25, dense, hybrid, and hybrid with domain metadata. BM25 gives the highest source recall (Recall@5 of 47%), and adding extracted metadata to hybrid retrieval does not improve it (39% versus 41%), which we attribute to noise in the extracted fields; the main bottleneck is locating the supporting span inside a correctly retrieved paper. As a materials-informatics application, literature-derived temperatures are matched to an independent theoretical reference set: 200 materials appear in both sources, and 20 have a median experimental critical temperature above the theoretical value.