EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Benchmarking Multimodal Large Language Models for Forensic Sciences: A Comprehensive Dataset and Evaluation Framework

Ashmaan Sohail, Om M. Patel, Jack C. S. Venditti, Jihwan Choi, Addison J. Wu

IEEE Access · 2026 · 14 · 68028-68038

DOI: 10.1109/access.2026.3684872 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Ashmaan Sohail und 4 weiteren Autor:innen. Die Arbeit nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Multimodal large language models (MLLMs) have demonstrated substantial progress in medical and legal domains in recent years; however, their capabilities from the lens of forensic science, a field at the intersection of complex medical reasoning and legal interpretation, remains largely unexplored. Forensic medicine uniquely depends on the accurate integration of often ambiguous text and visual information, yet systematic evaluations of MLLMs in this setting are lacking. We conducted a comprehensive benchmarking study of eleven state-of-the-art MLLMs, including proprietary (GPT-4o, Claude 4 Sonnet, Gemini 2.5 Flash) and open-source (Llama 4, Qwen 2.5-VL) models. Models were evaluated on 847 examination-style forensic questions drawn from various academic literature, case studies, and clinical assessments, covering nine forensic subdomains. Both text-only and image-based questions were included. Model performance was assessed using direct and chain-of-thought prompting, with automated scoring verified through manual revision. Performance improved consistently with newer model generations. Chain-of-thought prompting improved accuracy on text-based and multiple-choice tasks for most models, though this trend did not hold for image-based and open-ended questions. Visual reasoning and complex inference tasks revealed persistent limitations, with models underperforming in image interpretation and nuanced forensic scenarios. Model performance remained stable across forensic subdomains, suggesting topic type alone did not drive variability. These findings provide the first systematic baseline for MLLM performance in forensic science and inform pathways for their cautious integration into medico-legal workflows.

Zitieren

APA

Sohail, A., Patel, O. M., Venditti, J. C. S., Choi, J., & Wu, A. J. (2026). Benchmarking Multimodal Large Language Models for Forensic Sciences: A Comprehensive Dataset and Evaluation Framework. IEEE Access, 14, 68028-68038. https://doi.org/10.1109/access.2026.3684872

Vancouver

Sohail A, Patel OM, Venditti JCS, Choi J, Wu AJ. Benchmarking Multimodal Large Language Models for Forensic Sciences: A Comprehensive Dataset and Evaluation Framework. IEEE Access. 2026;14:68028-68038. doi:10.1109/access.2026.3684872

BibTeX
@article{Sohail2026,
  author = {Sohail, Ashmaan and Patel, Om M. and Venditti, Jack C. S. and Choi, Jihwan and Wu, Addison J.},
  title = {Benchmarking Multimodal Large Language Models for Forensic Sciences: A Comprehensive Dataset and Evaluation Framework},
  journal = {IEEE Access},
  year = {2026},
  volume = {14},
  pages = {68028--68038},
  doi = {10.1109/access.2026.3684872}
}
Kostenlos starten →