Papers by Baocai Shan
HuoziIME: An On-Device LLM-Enhanced Input Method for Deep Personalization (2026.acl-demo)
Copied to clipboard
| Challenge: | Mobile input method editors (IMEs) are the primary interface for text input, yet they are constrained to manual typing and struggle to produce personalized text. |
| Approach: | They propose a personalized on-device IME powered by large language models . they endow HUOZIIME with initial human-like prediction ability . |
| Outcome: | The proposed IME has initial human-like prediction ability and is optimized for on-device deployment. |
Dashboard2Code: Evaluating Multimodal Models on Reconstructing Interactive Dashboards (2026.acl-long)
Copied to clipboard
Tianhao Niu, Ziyu Han, Qiguang Chen, Shiqi Zhou, Baocai Shan, Hengjie Fang, Qingfu Zhu, Wanxiang Che
| Challenge: | Existing efforts to generate static visualizations focus on static charts and interactive dashboards. |
| Approach: | They propose a dashboard2code task that requires a model to explore an interactive dashboard, acquire feedback from its own interactions and generate code that reproduces the target dashboard. |
| Outcome: | The proposed task is based on 180 carefully designed and manually verified dashboard–code pairs spanning three difficulty levels and covering eight common real-world interaction patterns. |