Human-defined creativity is highly abstract, posing a challenge for multimodal large language models (MLLMs) to comprehend and assess creativity that aligns with human judgments. The absence of an existing benchmark further exacerbates this dilemma. To this end, we propose CreBench, which consists of two key components: 1) an evaluation benchmark covering the multiple dimensions from creative idea to process to products; 2) CreMIT (Creativity Multimodal Instruction Tuning dataset), a multimodal creativity evaluation dataset, consisting of 2.2K diverse-sourced multimodal data, 79.2K human feedbacks and 4.7M multityped instructions. Specifically, to ensure MLLMs can handle diverse creativity-related queries, we prompt GPT to refine the human feedback to activate stronger creativity assessment capabilities. CreBench serves as a foundation for building MLLMs that understand human-aligned creativity. Based on the CreBench, we fine-tune open-source general MLLMs, resulting in CreExpert, a multimodal creativity evaluation expert model. Extensive experiments demonstrate that the proposed CreExpert models achieve significantly better alignment with human creativity evaluation compared to state-ofthe-art MLLMs, including the most advanced GPT-4V and Gemini-Pro-Vision.

Diesel is the most consumed material in open-pit coal mines, and dump trucks are the equipment with the highest diesel consumption. In this study, the unit consumption of dump trucks was used as the predictor index, and the factors affecting the unit consumption of dump trucks (such as heavy load climbing height, unloaded distance, heavy load distance, heavy load weight, driver, temperature, rainfall, snowfall, wind speed, etc.) were used as the monitoring indicators. Based on historical data, the machine learning algorithm is used to train the unit consumption prediction model of dump trucks, and the unit consumption of a single dump truck is predicted in the future based on the current dump truck factor monitoring data and weather forecast data. The theoretical value of the unit consumption of the dump truck is calculated by using the historical data, the monitoring value of the unit consumption of the dump truck is calculated by real-time monitoring data, and the predicted value of the unit consumption of the dump truck is calculated through the model. By comparing the relationship between the theoretical value, the monitoring value and the predicted value, the analysis of whether the dump truck is abnormal fuel consumption for each transportation. If abnormal fuel consumption occurs, analyze the cause of abnormal fuel consumption, which is caused by conventional factors such as changes in the working area and weather changes, as well as oil leakage and other phenomena, so as to improve the management efficiency of relevant departments.