arXiv cs.AI· Marjan Celikik, Ana Peleteiro Ramallo, Javier Morales·· 9 天前AI 评分35
微调小模型超越前沿大模型：Qwen3.5 语法概念标注实现 16 倍成本降低
Grammar Concept Annotation at Scale: Deployed Fine-Tuned Small Language Models Outperform Prompted Frontier Models
AI 导读
研究团队微调 Qwen3.5 小语言模型完成语法概念标注，部署的 0.8B 模型在概念、证据跨度、正确性三项嵌套匹配标准下的精确率与召回率均超过 GPT-5.4 和 GPT-5.6 Sol，服务成本降低约 16 倍。在线实验显示学习者参与度提升 15.8%，新课程 GMV 增长 13.2%。
正文 · 原文
Abstract: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 ($+13.2\%$).
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| ACM classes: | I.2.7; I.2.6; K.3.1 |
| Cite as: | arXiv:2610.10827 [cs.CL] |
| (or arXiv:2610.10827v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.10827 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Marjan Celikik [view email]
[v1]
Wed, 7 Oct 2026 19:31:24 UTC (144 KB)
来源：arXiv cs.AI · arxiv.org