---
title: "TACT：面向教学自适应的英语辅导大模型后训练框架"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2608.03952v1"
permalink: "https://feed.astralor.com/items/itm_33751d14c59918d644d253bc"
published: "2026-08-04T17:16:14.000Z"
score: 77
channel: "tech"
tags: ["AI辅导","arXiv","Qwen3.5-4B","TACT","TACTCorpus","后训练"]
---

# TACT：面向教学自适应的英语辅导大模型后训练框架

## AI 点评

> 这篇论文的价值在于把教学理论真正落进了模型训练流程——用分类体系约束优化目标，而不再是靠对话流畅度打天下。对做AI教育产品、尤其是英语口语陪练的团队来说，TACT的开源数据与基准能省掉大量标注功夫，值得立刻上手试跑。

## 速览

TACT提出了一个基于人类教学研究框架的ESL辅导大模型后训练与评估方案。该框架构建了包含13种教师回应策略和学生学习行为分类的两套互补分类体系，并基于260段真实师生对话构建了带有32,379条标注的TACTCorpus。通过监督微调和分类对齐的GRPO优化，研究者训练出TACTutor模型，在TACTBench上相较骨干模型提升20.30%，并超越所有评估的专有基线。盲测中50名学习者给予其最高综合评分。数据和模型已开源。

## 关键点

- 提出双分类体系：13种教师回应策略和按动作类型与状态刻画的学生行为分类，为教学自适应提供结构化基础。
- 构建TACTCorpus，包含260段真实对话、32,379条人工标注及质量控制的增强训练数据。
- 基于Qwen3.5-4B进行SFT与分类对齐的GRPO后训练，TACTutor在TACTBench上提升20.30%，超越专有基线。
- 在50名学习者的盲评中获最高总体平均评分，并开源数据、基准与模型权重。

## 全文

Large language models (LLMs) are increasingly used to provide conversational practice for English-as-a-second-language (ESL) learners. Effective ESL tutoring, however, requires more than fluent response generation: a tutor must select an appropriate pedagogical action based on learner behavior and dialogue context. Human-tutoring research offers principles for adaptive support, but they are often task-specific and remain insufficiently integrated into LLM-based ESL tutor training and evaluation. We present TACT (Taxonomy-Aligned Conversational Tutor), a human-grounded framework for post-training and evaluating pedagogically adaptive ESL tutors. Drawing on established literature, we develop two complementary taxonomies: the Tutor-Strategy Taxonomy with 13 tutor response strategies and the Student-Move Taxonomy characterizing learner behavior by move type and status. Using these taxonomies, we construct TACTCorpus, which enriches 260 authentic teacher-student conversations with 32,379 annotations and quality-controlled augmented training data. We then post-train Qwen3.5-4B through supervised fine-tuning followed by taxonomy-aligned Group Relative Policy Optimization, producing TACTutor and optimizing it for scaffolding quality rather than reference imitation alone. On TACTBench, a strategy-balanced diagnostic benchmark comprising 78 authentic tutoring contexts, TACTutor improves over its backbone by 20.30% and outperforms all evaluated proprietary baselines under the same protocol, while maintaining backbone performance on established external educational benchmarks; in a blinded study with 50 learners, it also receives the highest overall mean rating among the evaluated tutors. We release the data, benchmark, and model weights, providing an open foundation for developing pedagogically adaptive ESL tutors.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_33751d14c59918d644d253bc