SUPR
TREND
Dnr:

NAISS 2025/22-321

Type:

NAISS Small Compute

Principal Investigator:

Jordanka Kovaceva

Affiliation:

Chalmers tekniska högskola

Start Date:

2025-03-07

End Date:

2026-04-01

Primary Classification:

10210: Artificial Intelligence

Webpage:

Allocation

Abstract

AI-Driven Analysis of Crash Narratives: Evaluating Open-Source LLMs for Structured Information Extraction The project aim is to develop and evaluate an open-source Large Language Model (LLM), trained on the NHTSA’s Crash Investigation Sampling System (CISS) dataset, capable of extracting missing details from textual crash reports. The primary objective is to assess the effectiveness of LLMs in understanding and interpreting accident narratives, enabling them to answer both binary and categorical questions about crash dynamics.