Open Category Submission

945 – Dual-Loop AI Writing Support Model: Bridging Creativity and Equity in Middle School Chinese Writing

This project introduces the Dual-Loop AI Writing Support Model, an innovative framework integrating generative AI into middle school Chinese writing instruction to address systemic gaps in traditional pedagogy. The model operates through two interconnected cycles are. Micro Loop and Macro Loop. Micro Loop is leveraging natural language processing (NLP), it provides real-time feedback on grammar, rhetoric, and style (e.g., correcting redundant plurals, suggesting vivid metaphors), enhancing textual accuracy and creativity. Macro Loop is utilizing learning analytics, it dynamically adapts teaching strategies (e.g., personalized prompts for struggling learners, nonlinear narrative tasks for advanced students) to foster equity and structural coherence.The Dual-Loop AI Writing Support Model revolutionizes middle school Chinese writing instruction by synergizing generative AI with pedagogical best practices. Its micro loop employs NLP-driven real-time feedback (e.g., grammar correction, metaphor suggestions) to refine textual accuracy and creativity, while the macro loop dynamically adapts teaching strategies using learning analytics (e.g., personalized prompts for struggling students, advanced tasks for high-achievers) to bridge equity gaps. Innovations include human-AI co-piloting (teachers design goals; AI handles granular tasks), culturally localized AI trained on Chinese rhetoric (e.g., 意境传达), and ethical safeguards (AI-disable protocols, revision tracking) to uphold integrity. Built on scalable cloud infrastructure (Tencent Cloud) and adaptive algorithms, the model demonstrated significant impact: experimental groups outperformed controls in narrative complexity (Cohen’s *d* = 0.67), while struggling learners improved structural coherence by 22%. By balancing efficiency, creativity, and equity, this framework offers a replicable blueprint for integrating AI into education—empowering teachers, engaging students, and aligning with global goals for ethical, sustainable EdTech innovation.

Amid rapid advancements in information technology, the global education sector is undergoing an unprecedented digital revolution. This transformation not only reshapes traditional teaching methods but also profoundly impacts educational content, pedagogy, and assessment systems. As the world’s second-largest economy, China’s educational digitization efforts are particularly noteworthy. The Chinese government has prioritized educational informatization, channeling policy guidance and funding to integrate educational technology with teaching practices, aiming to achieve modernization. However, in this era of sweeping digitization, the subject of Chinese language faces unique challenges and opportunities due to its cultural and humanistic essence. Chinese language education is not merely linguistic training but also a vital vehicle for cultural heritage, critical thinking, and aesthetic cultivation. Balancing the preservation of the discipline’s uniqueness with the effective use of digital technologies to enhance teaching outcomes remains a critical unresolved issue.
On one hand, digital technologies enrich Chinese language instruction with diverse resources and tools—such as multimedia courseware, online platforms, and intelligent grading systems—improving content delivery and efficiency. On the other hand, they challenge traditional pedagogical paradigms. Conventional Chinese teaching emphasizes teacher-led mentorship and student immersion through recitation, while digital pedagogy prioritizes student agency and interactivity, demanding new instructional philosophies and methods. Thus, the core question addressed by this research is: How can Chinese language education maintain its cultural and humanistic distinctiveness while innovatively leveraging digital technologies to achieve pedagogical advancement?

Technology & Resources:
Our solution leverages NLP tools (e.g., Transformer models) for real-time grammar/rhetoric analysis, dynamic lexicon matching for vocabulary suggestions, and a dual-column editor for traceable revisions. Development requires collaboration with AI engineers (for model fine-tuning), educators (for pedagogical alignment), and cloud computing resources (for scalable deployment). Pilot testing in Hong Kong schools (N=100 students) and partnerships with platforms like DouBao (a Chinese essay-grading AI) will validate technical feasibility.

Market Validation:
Demand is evidenced by China’s 2030 AI-in-education mandate and teacher surveys highlighting feedback delays (70% of educators report <24hr turnaround for essays). Pre-pilot data shows experimental groups outperformed controls in narrative complexity (d=0.67) and grammar accuracy (d=0.43). We will further validate demand via partnerships with regional education bureaus and EdTech conferences.

NA

Novelty:

Dual-Loop Model: Integrates immediate feedback (micro) with strategic adaptation (macro), addressing both surface errors (e.g., grammar) and deep challenges (e.g., creativity gaps). This surpasses single-loop AI tools like Grammarly.

Cultural Localization: AI trained on Chinese rhetoric (e.g., 意境传达) and ethical debates (e.g., “Human vs. AI Creativity”) ensures cultural relevance.

Human-AI Synergy: Unlike fully automated systems, our model positions teachers as “co-pilots,” preserving pedagogical agency while automating labor-intensive tasks (e.g., grammar checks).

Impact:

Creativity: Constraints like “3 sensory metaphors” force students to innovate beyond AI templates.

Equity: Struggling learners improved structural coherence by 22% via AI-generated scaffolds (e.g., narrative templates).

Ethics: “AI Disable Protocols” and revision tracking mitigate plagiarism risks, aligning with UNESCO’s AI ethics guidelines.

Scalability Strategies:

Modular Design: Components (e.g., NLP analyzer) can be adapted for other subjects (e.g., English writing) or regions via localization (e.g., training on Vietnamese metaphors).

Teacher Training: Workshops on AI prompt engineering and curriculum integration will ease adoption.

Cloud Infrastructure: Partnerships with Tencent Cloud ensure scalable server capacity.

Sustainability:

Environmental: Cloud-based deployment minimizes hardware waste.

User Engagement: Gamification (e.g., progress badges) and annual “AI Writing Olympics” sustain motivation.

Adaptability: Continuous model updates using learner behavior data (e.g., trending metaphors) ensure alignment with evolving curricula.

Long-Term Vision:
Lobby policymakers to integrate the model into national teacher training programs, ensuring systemic impact. Revenue from institutional licenses (e.g., school districts) will fund R&D for cross-modal AI (e.g., image-to-text tools).

The General Office of China’s Ministry of Education issued the Notice on Strengthening Artificial Intelligence (AI) Education in Primary and Secondary Schools in late last year, outlining six major tasks and initiatives: building a systematic AI curriculum, implementing regularized teaching and evaluation, developing universal teaching resources, constructing ubiquitous learning environments, promoting large-scale teacher training, and organizing diversified exchange activities. The notice explicitly sets the goal of universalizing AI education by 2030. Cities like Beijing and Shanghai have already launched regional action plans to deepen the integration of AI into subject-based teaching. This underscores the imperative for educators to prioritize the global wave of educational digitization.

Generative AI (e.g., ChatGPT, ERNIE Bot) has emerged as a core driver of educational transformation since its breakthrough in late 2022. Its strengths lie in generating high-quality text through large-scale corpus learning, enabling personalized feedback, multimodal interaction, and complex task reasoning. By 2024, generative AI in education has shifted from single-modality text processing to cross-modal integration, offering richer expressive support for writing instruction. Leveraging dynamic adaptation based on learning behavior data, AI can now provide differentiated support tailored to students’ varying abilities.

Personal Information Collection Statement (PICS):
1. The personal data collected in this form will be used for activity-organizing, record keeping and reporting only. The collected personal data will be purged within 6 years after the event.
2. Please note that it is obligatory to provide the personal data required.
3. Your personal data collected will be kept by the LTTC and will not be transferred to outside parties.
4. You have the right to request access to and correction of information held by us about you. If you wish to access or correct your personal data, please contact our staff at lttc@eduhk.hk.
5. The University’s Privacy Policy Statement can be access at https://www.eduhk.hk/en/privacy-policy.

• I have read and agree to the competition rules and privacy policy.