EN 登入 / 註冊

打造最強 AI 家教
重新定義未來的學習

Build the Ultimate AI Tutor
Redefining Future Learning

第一屆 Teaching Monster 挑戰賽:尋找能深刻理解人類學習需求、生成適性化影音教材的頂尖 AI Agent。

The 1st Teaching Monster Challenge: Seeking the top AI Agents capable of understanding learning needs and generating adaptive educational videos.

競賽宗旨 Mission Statement

「生成影片」不等於「有效教學」。

我們正在尋找具備 學科教學知識 (PCK) 的 AI:它不僅要懂知識,更要懂「如何教」。 從拆解概念、建立鷹架 (Scaffolding) 到差異化教學,挑戰 AI 在教育領域的極限。

"Generating video" is not "Effective Teaching".

We are looking for AI with Pedagogical Content Knowledge (PCK): An agent that understands not just the "What", but the "How". From deconstructing concepts and scaffolding to adaptive learning, we challenge the limits of AI in education.

挑戰任務與規則 Tasks & Rules

🎯 核心任務 🎯 Core Task

開發一組 AI Agent,透過 API 接收「學習需求」,全自動生成 人工智慧 / 機器學習 領域的英文教學影片。

  • 輸入:主題 (Topic) + 學生背景 (Persona)
  • 輸出:教學影片 (MP4) + 輔助教材
  • 限制:30分鐘內完成、全自動無人介入

Develop an AI Agent that takes "Learning Needs" via API and autonomously generates an English educational video on AI / Machine Learning topics.

  • Input: Topic + Student Persona
  • Output: Educational Video (MP4) + Supplementary Materials
  • Constraint: < 30 mins generation time, fully autonomous.

📝 評分標準 📝 Evaluation Criteria

由「AI 學生」與「人類學生」共同評比,決賽由大學教授親自驗收。

Judged by both "AI Students" and "Human Students", with finals reviewed by Professors.

40% 正確性Accuracy
零幻覺、引用規範 Zero Hallucination
30% 邏輯流暢Logic & Flow
教學鷹架、敘事連貫 Scaffolding, Coherence
20% 適配度Adaptability
感知學生程度 Context Awareness
10% 吸引力Engagement
多模態一致、有趣 Multimodal Delivery

賽程規劃 Timeline

3 月 1 日 - 熱身賽 March 1 - Warm-up

🚀 系統上線 & 壓力測試 🚀 Launch & Stress Test

釋出首波 20 題,由「AI 學生」提供自動化評測報告,協助團隊優化模型。

First 20 topics released. "AI Students" provide automated evaluation reports to help teams optimize models.

5 月 1 日 - 初賽啟動 May 1 - Preliminary

⚔️ 人類學生競技場 (Arena) ⚔️ Human Student Arena

進入 Blind Test 階段,由真實人類學生進行二選一評比 (Win/Loss),選出前三名。

Blind Test phase begins. Real human students vote on videos (Win/Loss) to select the top 3 teams.

6 月 12 日 - 決賽巔峰 June 12 - Grand Final

🏆 教授專家總驗收 🏆 Professor Review

頂尖教授出題,限時挑戰高難度教學場景。最終於 6/26 舉辦成果發表會。

Top experts pose challenging teaching scenarios. Final winners announced at the workshop on June 26.

主辦單位 Organizer

台灣大學人工智慧頂尖研究中心 (NTU AI-CoRE) National Taiwan University AI Center of Research Excellence (NTU AI-CoRE)

聯絡我們 Contact Us