Large Language Models (LLMs) have shown increasing potential in automating model-driven software engineering tasks, particularly in generating models conforming to Domain Specific Languages (DSLs) from natural language. While most existing approaches rely on large proprietary models, their high cost and limited deployability hinder broader adoption. In this paper, we evaluate whether open-source LLMs of varying sizes (0.5B to 32B parameters) can generate DSL-conformant models using only few-shot prompting, without any fine-tuning. Our evaluation focuses on key model-driven engineering (MDE) requirements, including syntactic validity, semantic completeness, and inter-model reference consistency. We extend our prior work by moving from generating user interface models (referred to as “UI models” in this paper) over fixed, predefined data schemas (“data models”) to generating both the UI and data models entirely from scratch. This shift serves two purposes: first, it highlights the LLM’s ability to infer domain-specific relationships and maintain consistency across multiple interconnected models; second, it allows us to generalize earlier findings by testing DSL generation across models of different natures and structural roles. Our structured evaluation combines automatic parsing and expert feedback across 39 LLMs, revealing that several compact models (e.g., gemma3:12b, mistral:7b-instruct) approach or match the quality of much larger models. These findings demonstrate the feasibility of using smaller, open-source LLMs for grammar-conformant DSL generation in MDE workflows, offering a cost-effective and deployable alternative to closed LLMs.

This research introduces an innovative framework for the prediction of human hand motion in a shared workspace with a robot, fostering safe and efficient human-robot collab-oration. Due to the absence of benchmark datasets for this task, we created a custom dataset of human hand trajectories by orchestrating intentional collisions between humans and robots during data collection. To enable efficient processing and prediction, our framework leverages the quantization of sensitive human hand positions into small 3D cells. These cells are later modeled for learning the embedding for better human hand motion prediction. Notably, our framework outperforms the baseline model. Al-though the enhanced predictive power entails extra computation for finding the Nearest Neighbors (NN) during quantization, we efficiently manage this cost through the integration of off-the-shelf information retrieval frameworks like ANNOY. This strategic approach ensures real-time performance and maintains precise approximate NN results.

The advent of 3D LiDAR technology has revolutionized object detection in applications such as autonomous driving, robotics, and advanced driver assistance systems. However, existing methods often require substantial computational resources, limiting their practicality for real-time applications on devices with constrained hardware capabilities. This paper presents an efficient and lightweight 3D LiDAR object detection framework, 3D-PSH, that combines adaptive clustering with 3D Point Spatial Histograms (3D-PSH) and classical classification techniques to address these challenges. Our framework begins with an adaptive clustering algorithm that segments the point cloud data into distinct clusters, representing potential objects. 3D Point Spatial Histograms (3D-PSH) are then computed from these clusters and subsequently quantized into a Bag of Visual Words (BoVW) to create a compact and informative representation. These representations are then classified using robust classical classification methods to identify object types, such as pedestrians and vehicles. This multi-step approach ensures a balance between computational efficiency and detection accuracy, making it suitable for real-time deployment. Extensive experiments on the KITTI dataset and our live sensor data demonstrate the effectiveness and efficiency of our proposed framework. The results indicate that our method achieves competitive accuracy while significantly reducing computational requirements compared to traditional approaches. This framework offers a practical solution for deploying 3D object detection in a wide range of applications, particularly where computational resources are limited.

Rapid increase in video databases has forced the industry to have efficient and effective frameworks for video retrieval and indexing. Video segmentation into scenes is widely used for video summarization, partitioning, indexing and retrieval. In this paper, we propose a framework for scene detection mainly based on entropy and Speeded Up Robust Features (SURF) features. First, we detect the fade and abrupt boundaries based on frame entropy analysis and SURF features matching. Fade boundaries are smart indication of scenes beginning or ending in many videos and dramas, and are detected by frame entropy analysis. Before abrupt boundary detection, unnecessary frames which are obviously not abrupt boundaries, such as blank screens, high intensity influenced images, sliding credits, are removed. Candidate boundaries are detected to make SURF features efficient for abrupt boundary detection, and SURF features between candidate boundaries and their adjacent frames are used to detect the abrupt boundaries. Second, key frames are extracted from abrupt shots. We evaluate our key frame extraction with other famous algorithms and show the effectiveness of the key frames. Finally, scene boundaries are detected using sliding window of size K over the key frames in temporal order. In experimental evaluation on the TRECVID-2007 shot boundary test set, the algorithm for shot boundary achieves substantial improvements over state-of-the-art methods with the precision of 99% and the recall of 97.8%. Experimental results for video segmentation into scenes are also promising, compared to famous state-of-the-art techniques.