← 返回研究新聞
寫作教師與兩名大學生在研討桌旁對照帶批註的段落修改稿,簡潔的視覺連線呈現不同階段的修改重點。
會議論文會議論文20262026年10月23日· 3 min

會議論文:FeedType 檢驗 AI 寫作反饋能否隨修改過程調整重點

Norah Almousa, Shayan Peyghambari Oskoui, Raquel Coelho, Gayle Rogers, Xiang Lorraine Li, Diane Litman

Proceedings of the Artificial Intelligence in Measurement and Education Conference (AIME-Con): Full Papers

500 字摘要

寫作教師與兩名大學生在研討桌旁對照帶批註的段落修改稿,簡潔的視覺連線呈現不同階段的修改重點。

Norah Almousa 及其同事在 10 月 5–7 日舉行的 AIME-Con 會議論文集中發表 FeedType。基準使用三門大學寫作課程的 1,966 個段落,在三種提示策略下比較教師反饋和六種語言模型,並以七類內容描述反饋重點。自動標註經過人工判斷核對。模型雖然覆蓋了大多數類別,卻沒有完全再現教師的適應性模式。修改階段與表現水平的分析僅限寫作課程中的 composition 課程。論文考察反饋行為,沒有檢驗干預對學習的影響。

這項研究提供了一種提問方式:評語針對什麼問題,其重點是否會適當改變?即使另一種回應更有用,系統也可能反覆要求學生補充細節。即使模型的類別分佈與教師接近,也不能單憑這一點證明每條評語準確或有幫助。課程、標註選擇和受測模型版本,同樣限制了結論適用的範圍。

對 AIEDHK 讀者而言,可考慮圍繞同一段落的兩版草稿開展寫作工作坊。閲讀生成反饋前,教師先判斷當前作業階段最重要的下一步。學生隨後比較建議,找出每條建議針對的文字證據,並解釋為什麼某個問題應優先修改。這是本站提出的教學設計,並非基準研究已經評估的活動。

例如,關於本地環境議題的初稿可能需要更清楚的主張和適當來源;後續稿件也許已解決這項問題,轉而需要精確説明一項限制條件。兩次都要求“增加細節”,就可能遺漏不同階段的修改重點。學生可解釋為什麼一條表面有用的建議會分散當前任務的注意力,讓教師看到其寫作判斷。

本地試點可保留草稿、建議、學生選擇的行動及簡短理由。教師抽查有分歧和無分歧的案例,而不應假定接受更多建議就代表學習更好。審閲也應包含類別判斷正確、但修改方案不適當的評語。目標是瞭解反饋如何支持寫作過程中特定時刻的決策。

雙語寫作課還可檢查評語是否在跨語言表達時保留作者原意。學生可依據作業目的解釋修改,教師則考察語言潤色是否掩蓋了尚未解決的論證或缺乏依據的主張。

產品團隊可在發佈前明確一個評估問題:反饋是否回應了作業、當前草稿和學生表現出的困難?這能指導範圍明確的審閲樣本,又不必把教師過去的做法當作不可質疑的標準。之後減少輔助的獨立寫作可提供補充學習證據。FeedType 是檢查反饋適應性的研究資源;課堂或機構採用系統,仍需瞭解學生實際收到的評語質量及其後果。

相關論文

Illustrative research-review scene with two raters comparing repeated and progressive activity sequences beside a screen with one rising and one falling chart.
工具 / 數據集2026
工具 / 數據集 239

Preprint: Raising a tutor metric lowered human ratings in an optimization experiment

Daniel Domínguez Figaredo, Rafael Fernández De la Cruz

arXiv preprint

An optimization experiment produced higher automated scores but lower human ratings for instructional sequences. Its synthetic cases and constrained intervention make it a warning about evaluation design, not evidence of harm to actual students.

preprintAI tutoringevaluation validity
閱讀 500 字摘要 →
A South Asian female lecturer and a White male university student examine two differently annotated statistics feedback sheets beside a hand-drawn confidence-interval diagram
期刊論文2026
期刊論文 216

Students’ feedback ratings reflected source credibility without consistently matching expert judgments

Tanya Nazaretsky, Hagit Gabbay, Tanja Käser

Computers and Education: Artificial Intelligence

Researchers compared human and AI feedback on 979 responses from six university STEM courses, then examined ratings from 472 students. Expert analysis found broadly similar pedagogical quality, while students’ judgments were associated with their views of the source’s credibility. The study did not measure feedback uptake or learning gains. For educators, it raises a practical question: how can students justify an evaluation of advice before deciding whether to use it?

feedback literacysource credibilityhigher education
閱讀 500 字摘要 →
A mathematics lecturer and peer discuss a university student's induction proof beside a laptop showing feedback on the next step.
會議論文2026
會議論文 206

LeanTutor: Towards a Verified AI Mathematical Proof Tutor

Manooshree Patel, Rayna Bhattacharyya, Thomas Lu, Arnav Mehta, Niels Voss, Narges Norouzi, Gireeja Ranade

Proceedings of the AAAI Conference on Artificial Intelligence

LeanTutor connects a language model to the Lean theorem prover to support mathematical proof feedback. Its technical evaluation shows promising hints on a small selected benchmark, but also exposes a crucial boundary: a verifier can check the formal proof only after the learner's words have been translated correctly. The study did not test classroom learning, and feedback evaluation excluded failed translations. It is an early system design with explicit limitations, not a validated replacement for mathematics instruction.

Mathematics EducationIntelligent TutoringFormal Verification
閱讀 500 字摘要 →