We introduce a new benchmark, ChartMimic, aimed at assessing the visually-grounded code generation capabilities of large multimodal models (LMMs). ChartMimic utilizes information-intensive visual charts and textual instructions as inputs, requiring LMMs to generate the corresponding code for chart rendering. ChartMimic includes $4,800$ human-curated (figure, instruction, code) triplets, which represent the authentic chart use cases found in scientific papers across various domains (e.g., Physics, Computer Science, Economics, etc). These charts span $18$ regular types and $4$ advanced types, diversifying into $201$ subcategories. Furthermore, we propose multi-level evaluation metrics to provide an automatic and thorough assessment of the output code and the rendered charts. Unlike existing code generation benchmarks, ChartMimic places emphasis on evaluating LMMs' capacity to harmonize a blend of cognitive capabilities, encompassing visual understanding, code generation, and cross-modal reasoning. The evaluation of $3$ proprietary models and $14$ open-weight models highlights the substantial challenges posed by ChartMimic. Even the advanced GPT-4o, InternVL2-Llama3-76B only achieved an average score across Direct Mimic and Customized Mimic tasks of $82.2$ and $61.6$, respectively, indicating significant room for improvement. We anticipate that ChartMimic will inspire the development of LMMs, advancing the pursuit of artificial general intelligence.

Urban migrant children encounter difficulties in developing a sense of belonging, which compromises their living experiences and academic performance. This paper introduces Community Tour, an expandable knowledge exploration system designed to assist migrant children with their extracurricular learning, and empower social workers to systematically carry out community events. A knowledge exploration interaction process has been designed to localize STEAM education with community elements through practical tasks, learning motivation, and achievements. Prototypes of interactive installation and back-end platform have been built and partial validation experiments have been conducted, with future work focusing on collaborating with communities for field testing and design iteration. The sustainability of the system lies in the potential for education on various themes, and its compatibility from urban villages to more regular communities, contributing to the child-friendly cities.

Cellular Trending is an artwork that reveals information fragmentation on the social media through generative lens. It visualizes the fragmented information dissemination in social media with affective attributes integrated cellular automata to create artistic experience. A multi-level interactive system consists of CELL, FACT and VIEW is proposed based on information dissemination theory mapping to fragmented communication, thinking and reading. From information acquisition to opinion expression, the artwork resonates and arouses people's reflection on modern information dissemination and information acquisition behavior in the digital age.

"AI mirror", an interactive art, tends to visualize the self-knowledge mechanism from the AI's perspective, and arouses people's reflection on artificial intelligence. In the first stage of the unconscious imitation, the visual neurons perceive environmental information and mirror neurons imitate human behavior. Then, the language and consciousness are generated from the long term of imitation, denoted as poet and coordinates in an affective space. In the final stage of conscious behavior, an affinity analysis is generated, and the mirror neurons will behave more harmoniously with the user or have the autonomous movements on its own, which evokes the user's reflection on its undiscovered traits.