Evaluating Semantic and Syntactic Understanding in Large Language Models for Payroll Systems Journal: Springer Journal
Hendrika Maclean, Mert Can Cakmak, Muzakkiruddin Ahmed Mohammed, Shames Al Mandalawi, John Talburt · 2026
Read Evaluating Semantic and Syntactic Understanding in Large Language Models for Payroll Systems Journal: Springer Journal by Hendrika Maclean, Mert Can Cakmak, Muzakkiruddin Ahmed Mohammed, Shames Al Mandalawi, John Talburt (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-2601-18012.txt.
Description
Hendrika Maclean and co-authors test language models on the strict world of payroll. Their 2026 preprint notes a known weakness. Models are strong at writing, search, and analysis, and their grasp of language keeps growing. Yet they stay shaky on exact numbers and on output that auditors can check line by line. Payroll needs both: exact math and a clean audit trail.
The team studies how well models handle the semantic and syntactic sides of payroll tasks. Meaning and form both matter when pay rules are coded in words. Errors here hit real wages and real legal duties. The findings help firms judge where a model can assist payroll work and where humans must stay in charge.
Catalog record
| CREATOR | Hendrika Maclean, Mert Can Cakmak, Muzakkiruddin Ahmed Mohammed, Shames Al Mandalawi, John Talburt |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 25 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-18012