TY - JOUR T1 - Prompt perturbation and fraction facilitation sometimes strengthen large language model scores JO - Data and Information Management UR - https://eprints.whiterose.ac.uk/id/eprint/240989 PY - 2026/05/08 AU - Thelwall M ED - DO - DOI: 10.1016/j.dim.2026.100140 PB - Elsevier BV VL - 10 IS - 3 Y2 - 2026/05/13 ER -