Extending Confidence-Based Text2Cypher with Grammar and Schema Aware Filtering
Makbule Gulcin Ozsoy · 2026
Read Extending Confidence-Based Text2Cypher with Grammar and Schema Aware Filtering by Makbule Gulcin Ozsoy (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-10318.txt.
Description
Makbule Gulcin Ozsoy studies how to make language models query graph databases more reliably. Large language models can turn plain questions into Cypher, the query language for graph databases. Yet made-up queries often break syntax rules or clash with the database schema. Her 2026 paper tests structural checks applied at run time, after generation.
Ozsoy extends a confidence-based framework with step-by-step filtering. Candidate queries pass through confidence scoring, grammar checks, and schema checks before final pooling. Tests with two tuned models showed that grammar checks improved valid syntax, while schema checks further improved run quality by enforcing fit with the database structure. Stronger filtering also raised blank predictions and reduced coverage, a trade-off the paper maps clearly.
Catalog record
| CREATOR | Makbule Gulcin Ozsoy |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 27 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.10318