Computer-Assisted Language Learning

Follow this topic Ask about this topic

Latest papers

9 papers1 paper in the last four weeks, against 2 the four weeks before. 0.0% of all new papers.

Grammar Concept Annotation at Scale: Deployed Fine-Tuned Small Language Models Outperform Prompted Frontier Models
Oct 7, 2026Marjan Celikik, Ana Peleteiro Ramallo, Javier MoralesSmall Language Models Computer-Assisted Language Learning
Preply
LLMersion: A Local-First AI Agent Framework for Low-Cost Home Language Learning toward Educational Equity
Sep 12, 2026Qiming Guo, Jinwen Tang, Xingran Huang +3Computer-Assisted Language Learning AI in Education
Dept. of Computing Sciences Texas A&M University–Corpus Christi Corpus Christi, TX, USA · Dept. of Electrical Engineering and Computer Science University of Missouri Columbia, MO, USA · Dept. of Computer Science and Engineering University of California, Riverside Riverside, CA, USA +2
Structure and Implementation of New Practical English Textbooks Driven by Artificial Intelligence
Sep 2, 2026Ya Wang, Lei Zhang, Xueguang Yang +1Intelligent Tutoring Systems AI in Education
Air Force Logistics Academy, Xuzhou, 221000,China
Question Type, Cognitive Load, and CEFR Alignment: Evaluating LLM-Generated EFL Grammar Drill Exercises
Jun 1, 2026Steve Woollaston, Brendan Flanagan, Yuko Toyokawa +1LLM Evaluation Computer-Assisted Language Learning
Kyoto University, Japan · Ritsumeikan University, Japan
Classification of non-analyzable word types in web documents to implement an effective Korean e-learning system
May 28, 2026Sang-Taek Park, Ae-Lim Ahn, Eric Laporte +1Corpus Linguistics Computer-Assisted Language Learning
DICORA, Hankuk University of Foreign Studies, Korea · LIGM, Université Paris-Est, France
Controllable Spoken Dialogue Generation: An LLM-Driven Grading System for K-12 Non-Native English Learners
Apr 24, 2026Haidong Yuan, Haokun Zhao, Wanshi Xu +4RL for Language Models LLM Alignment
Peking University · Fudan University · Tencent +2
AFRILANGTUTOR: Advancing Language Tutoring and Culture Education in Low-Resource Languages with Large Language Models
Apr 22, 2026Tadesse Destaw Belay, Shahriar Kabir Nahin, Israel Abebe Azime +6Multilingual Language Models Intelligent Tutoring Systems
Instituto Politécnico Nacional, Mexico · University of South Florida, FL, USA · Saarland University, Germany +2
Listen, Correct, and Feed Back: Spoken Pedagogical Feedback Generation
Mar 28, 2026Junhong Liang, Yifan Lu, Ekaterina Kochmar +1Computer-Assisted Language Learning Educational Feedback Generation
Mohamed bin Zayed University of Artificial Intelligence, Abu Dhabi, UAE
Profiling learners' affective engagement: Emotion AI, intercultural pragmatics, and language learning
Mar 20, 2026Robert Godwin-JonesComputer-Assisted Language Learning Affective Computing
Virginia Commonwealth University

Oct 7, 2026J/K move · Enter open · S save

Grammar Concept Annotation at Scale: Deployed Fine-Tuned Small Language Models Outperform Prompted Frontier Models

Marjan Celikik, Ana Peleteiro Ramallo, Javier Morales

Preply

Small Language Models Computer-Assisted Language Learning Language Model Post-Training LLM Evaluation

Open paper AskPDF arXiv

Corrective feedback is among the best-evidenced drivers of second-language acquisition, yet corrections delivered during lessons rarely accumulate into an actionable view of grammar mastery. Prompted frontier models can provide such a view from learner--tutor lesson transcripts, but they are costly at scale. We close this gap by fine-tuning Qwen3.5 small language models (SLMs) on filtered and rebalanced teacher-generated supervision, then deploying an efficient 0.8B model in an end-to-end grammar mastery tracker for all English learners on our platform. Internalizing the annotation contract into adapter weights enables pairing the 0.8B model with a compact matched prompt rather than verbose instructions. On two human-curated benchmarks, both the deployed 0.8B model and a 4B reference comparator outperform prompted GPT-5.4 and GPT-5.6 Sol in precision and recall under nested matching criteria of increasing strictness: concept, evidence span, and correctness. The deployed 0.8B SLM reduces serving cost by approximately 16

\times

. A feature-level online experiment shows significant gains in learner engagement (

+15.8\%

) and key business metrics, including scheduled hours (

+2.1\%

) and GMV from new lessons (

+13.2\%