Traffic-IT: Enhancing traffic scene understanding for multi

designed to enhance MLLMs’ capabilities in traffic scene understanding. The dataset covers various traffic scenarios， Traffic-IT significantly improves MLLMs’ performance in interpreting complex traffic scenes. We anticipate that Traffic-IT will be a crucial resource for future developments in smart city applications. Record URL: Availability: Supplemental Notes: © 2025 Elsevier Ltd. All rights are reserved， yet they are typically trained on general datasets， AI training，950 question-and-answer pairs from 30， providing in-depth insights and driving strategies tailored to real-world needs. Created through expert consultation and rigorous validation， Traffic-IT: Enhancing traffic scene understanding for multimodal large language models In recent years， traditional models often lack the generalizability needed to adapt to diverse traffic scenarios. Multimodal large language models (MLLMs) offer a promising solution， limiting their effectiveness in specific transportation contexts. To address this，000 images， and similar technologies. Abstract reprinted with permission of Elsevier. The contents of this paper reflect the views of the author[s] and do not necessarily reflect the official views or policies of the Transportation Research Board or the National Academy of Sciences. Authors: Publication Date: 2025-11 Language English Media Info Subject/Index Terms Filing Info ， including weather conditions， and times of day， we introduce Traffic-IT， locations， a dataset comprising 220， including those for text and data mining， the convergence of artificial intelligence and urban infrastructure has driven transformative advances in intelligent transportation systems (ITS). However，。