学生数据悖论与单学生-导师模型的奇特案例：为个性化学习训练大语言模型的退行性副作用
Student Data Paradox and Curious Case of Single Student-Tutor Model: Regressive Side Effects of Training LLMs for Personalized Learning
AI总结：
本研究揭示了为个性化学习训练LLM时出现的“学生数据悖论”——在学生对话数据上训练会损害模型的事实知识与推理能力，通过多基准测试验证了该现象，并提出幻觉令牌缓解策略，指出需平衡学生行为建模与模型教育完整性。
摘要：
对个性化教育的追求推动了大语言模型（LLMs）融入智能辅导系统的开发过程。为了更好地理解并适配学生的个体需求（包括他们的错误认知），LLMs需要在海量的学生-导师对话数据集上进行训练。本研究揭示了该方法中存在的一个根本性挑战：“学生数据悖论（Student Data Paradox）”。当为理解学习者行为而在学生数据上训练的LLMs无意中损害了自身的事实知识与推理能力时，这一悖论便会显现。\n我们通过在学生-导师对话数据集上训练前沿语言模型，并在多个基准测试上评估其性能，对该悖论展开了研究。这些基准测试评估了语言模型能力的多个维度，包括推理能力、真实性以及常识理解。研究结果显示，模型在这些不同基准测试上的性能均出现显著下降，表明当模型被训练用于建模学生行为时，其能力受到了广泛影响。\n本研究主要有两项贡献：（1）通过对模型性能的定量分析，实证证明了学生数据悖论的存在；（2）提出“幻觉令牌（hallucination tokens）”作为缓解策略。这些令牌在提升性能的同时，也凸显了在精准的学生行为建模与维持LLM作为教育工具的完整性之间取得平衡的长期挑战。本研究强调，亟需创新解决方案来调和两大目标的冲突：既要忠实理解多样化的学生认知，又要保留模型提供准确信息与指导的能力。
英文摘要：
The pursuit of personalized education has led to the integration of Large Language Models (LLMs) in developing intelligent tutoring systems. To better understand and adapt to individual student needs, including their misconceptions, LLMs need to be trained on extensive datasets of student-tutor dialogues. Our research uncovers a fundamental challenge in this approach: the ``Student Data Paradox.'' This paradox emerges when LLMs, trained on student data to understand learner behavior, inadvertently compromise their own factual knowledge and reasoning abilities. We investigate this paradox by training state-of-the-art language models on student-tutor dialogue datasets and evaluating their performance across multiple benchmarks. These benchmarks assess various aspects of language model capabilities, including reasoning, truthfulness, and common sense understanding. Our findings reveal significant declines in the models' performance across these diverse benchmarks, indicating a broad impact on their capabilities when trained to model student behavior. Our research makes two primary contributions: (1) empirical demonstration of the Student Data Paradox through quantitative analysis of model performance, and (2) introduction of ``hallucination tokens'' as a mitigation strategy. These tokens, while improving performance, highlight the persistent challenge of balancing accurate student behavior modeling with maintaining the LLM's integrity as an educational tool. This study emphasizes the need for innovative solutions to reconcile the conflicting goals of faithfully understanding diverse student cognition while preserving the model's ability to provide accurate information and guidance.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
