Hongliang Zhong

Zhijun Chen, Zhe Lu, Qiushi Chen et al.

Short-term traffic flow prediction is a vital branch of the Intelligent Traffic System (ITS) and plays an important role in traffic management. Graph convolution network (GCN) is widely used in traffic prediction models to better deal with the graphical structure data of road networks. However, the influence weights among different road sections are usually distinct in real life, and hard to be manually analyzed. Traditional GCN mechanism, relying on manually-set adjacency matrix, is unable to dynamically learn such spatial pattern during the training. To deal with this drawback, this paper proposes a novel location graph convolutional network (Location-GCN). Location-GCN solves this problem by adding a new learnable matrix into the GCN mechanism, using the absolute value of this matrix to represent the distinct influence levels among different nodes. Then, long short-term memory (LSTM) is employed in the proposed traffic prediction model. Moreover, Trigonometric function encoding is used in this study to enable the short-term input sequence to convey the long-term periodical information. Ultimately, the proposed model is compared with the baseline models and evaluated on two real word traffic flow datasets. The results show our model is more accurate and robust on both datasets than other representative traffic prediction models.

Can Wang, Hongliang Zhong, Menglei Chai et al.

Automatic furniture layout is long desired for convenient interior design. Leveraging the remarkable visual reasoning capabilities of multimodal large language models (MLLMs), recent methods address layout generation in a static manner, lacking the feedback-driven refinement essential for interactive user engagement. We introduce Chat2Layout, a novel interactive furniture layout generation system that extends the functionality of MLLMs into the realm of interactive layout design. To achieve this, we establish a unified vision-question paradigm for in-context learning, enabling seamless communication with MLLMs to steer their behavior without altering model weights. Within this framework, we present a novel training-free visual prompting mechanism. This involves a visual-text prompting technique that assist MLLMs in reasoning about plausible layout plans, followed by an Offline-to-Online search (O2O-Search) method, which automatically identifies the minimal set of informative references to provide exemplars for visual-text prompting. By employing an agent system with MLLMs as the core controller, we enable bidirectional interaction. The agent not only comprehends the 3D environment and user requirements through linguistic and visual perception but also plans tasks and reasons about actions to generate and arrange furniture within the virtual space. Furthermore, the agent iteratively updates based on visual feedback from execution results. Experimental results demonstrate that our approach facilitates language-interactive generation and arrangement for diverse and complex 3D furniture.