TACT: Taxonomy-Aligned Post-Training for Pedagogically Adaptive English Tutoring

2026-08-04

AI Summary

This study addresses the limited pedagogical adaptability of current large language models in English as a Second Language (ESL) tutoring, which often lack grounding in established instructional theory. To bridge this gap, the authors propose TACTutor, a novel system that first establishes a dual-layer taxonomy aligning teacher strategies with student behaviors, enabling the creation of TACTCorpus—a richly annotated real-world dialogue dataset. Building upon this foundation, they introduce Group Relative Policy Optimization, a post-training approach combining supervised fine-tuning with classification-aligned reward modeling to refine Qwen3.5-4B. This work pioneers a computationally tractable framework for encoding human tutoring strategies, marking a significant shift from mere imitation to scaffolded instructional quality. Evaluated on the newly developed diagnostic benchmark TACTBench, TACTutor outperforms all baselines by 20.30%, surpasses all closed-source models, achieves the highest average rating in blind tests with 50 learners, and maintains strong performance on external educational benchmarks.

Abstract

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.

Problem

• pedagogical adaptation
• ESL tutoring
• taxonomy-aligned evaluation
• adaptive support
• conversational tutoring

Innovation

• taxonomy-aligned training
• pedagogical adaptation
• scaffolding optimization
• Group Relative Policy Optimization
• ESL tutoring