default author photo

Jordanka Kovaceva

Chalmers University of Technology

SCHOLARLY PAPERS

2

DOWNLOADS

50

TOTAL CITATIONS

0

Scholarly Papers (2)

Actions:
Email selected abstracts
1.

Domain-Adapted Pre-trained Language Models for Implicit Information Extraction in Crash Narratives

Number of pages: 45 Posted: 11 Oct 2025 Last Revised: 15 Oct 2025
Xixi Wang, Jordanka Kovaceva, Miguel Costa, Shuai Wang, Francisco Camara Pereira and Robert Thomson
Technical University of Denmark, Chalmers University of Technology, Technical University of Denmark, affiliation not provided to SSRN, Technical University of Denmark and Chalmers University of Technology
Downloads 43 (1,145,425)

Abstract:

Loading...

Traffic Safety, Crash Narratives, Pre-trained Language Models, Fine-Tuning, Information Extraction

2.

Where and Why Predictions Fail: Uncertainty-Decomposition-Based Failure Diagnosisfor Crash Injury Severity

Number of pages: 31 Posted: 29 May 2026
Junhao Wei, Fusako Sato, Robert Thomson, Jordanka Kovaceva and Yusuke Miyazaki
Institute of Science Tokyo, Japan Automobile Research Institute, Chalmers University of Technology, Chalmers University of Technology and Institute of Science Tokyo
Downloads 7 (1,575,921)

Abstract:

Loading...

injury prediction, crash analysis, failure-mode diagnosis, Safety governance, Uncertainty Quantification