Designing project-based learning (PBL) demands managing highly interdependent components, a task that both traditional linear tools and purely conversational AI struggle with. Traditional tools fail to capture the non-linear nature of creative design, while conversational systems lack the persistent, shared context necessary for reflective collaboration. Grounded in theories of distributed cognition, we introduce CoMAP, a system that embodies a graph-based collaboration paradigm. By providing a shared visual workspace with dual-modality AI support, CoMAP transforms the human-AI relationship from a prompt-and-response loop into a transparent and equitable partnership. Our study with 30 educators shows CoMAP significantly improves teachers’ design expression, divergent thinking, and iterative practice compared to a dialogue-only baseline. These findings demonstrate how a nonlinear, artifact-centric approach can foster trust, reduce cognitive load, and support educators to take control of their creative process. Our contributions are available at: https://comap2025.github.io/.

With the rapid development of multimodal machine learning, data security has become a critical bottleneck constraining multimodal artificial intelligence advancement. Multimodal unlearnable examples, as an emerging data protection paradigm, prevent unauthorized model training by disrupting cross-modal semantic consistency through adversarial perturbations and trigger identifiers. However, existing multimodal protection mechanisms suffer from significant robustness deficiencies and security vulnerabilities. This paper conducts a systematic security evaluation of multimodal unlearnable examples and proposes a Cross-Modal Collaborative Recovery (CMCR) attack framework. CMCR integrates Joint Conditional Diffusion Purification for image denoising, Adaptive Text Purification for trigger removal, and Visual Cross-modal BART for cross-modal reconstruction. Experiments demonstrate that CMCR successfully restores image-text retrieval performance to levels approaching clean data across different model architectures, achieving an average recovery rate of 89.0%. Our results reveal that current multimodal unlearnable examples have security vulnerabilities and insufficient robustness against coordinated cross-modal attacks.

Industrial Internet of things (IIoT) is a typical application scenario in the sixth generation (6G) mobile networks. IIoT scenarios involve dense multipath components (MPCs) and nonnegligible scattering components caused by many moving objects. In this paper, image method (IM) is applied and extended to analyze the channel properties of IIoT. Directive model is modified to adapt to IM. The moving patterns of objects are defined and their snapshots are established along the time axis. Multiple-input multiple-output (MIMO) is supported as it is widely applied in IIoT. A smart warehouse scenario equipped with moving handcars is selected to analyze the channel of IIoT scenario. Parameters such as azimuth angle, elevation angle, angular spread, power, and delay spread of received rays are calculated and compared with those generated by quasi-deterministic (Q-D) model traditionally used in IM. Maximum and minimum Doppler shifts, received power, and delay spread are calculated along the time axis to analyze the influence of mobility to channel properties. The results show that directive model generates scattering components more realistically compared with Q-D model, and that the channel properties may experience sudden changes due to the line-of-sight (LoS) component being obstructed.

In this paper, computation reduction algorithm of motion estimation is developed for MPEG4 video encoders. Our focus is on the computation reduction for block based integer and half pixel search. A method is proposed to reduce the search points and computation with little extra tradeoff and small quality downgrade. In MPEG4, the motion estimation is mainly composed of three major parts: 1) The Macroblock based integer pixel search. 2) The block based integer pixel search. 3) The half pixel search. The Macroblock based integer pixel search usually consumes the most computation power. Therefore many fast search algorithms are employed instead of the full search. In contrast, as to 2), 3), since the search range is quite small, usually the full search method is used. With many novel algorithms the computation requirements in 1) decrease greatly. Hence, the percentage of 2) and 3) are increasing and come to an important position for computation reduction. The method proposed in this paper jointly considered the ME, DCT and quantization. It takes advantage of the idea in both MB and block layer. The computation reduction is obvious from experiment. Because the full search algorithm is still in use, the quality downgrade is little compared to the old computation reduction algorithms.