将LLM标注用于BERT分类器：风险自负
Feeding LLM Annotations to BERT Classifiers at Your Own Risk
AI总结：
本研究实证分析LLM生成标签微调BERT分类器的风险，发现性能下降、不稳定及过早平台期，提出熵过滤与集成缓解策略，但警示高风险任务需谨慎。
摘要：
使用LLM生成的标签来微调较小的仅编码器模型以进行文本分类，在各种场景中已变得流行。尽管这种方法在简单且低风险的应用中可能是合理的，我们通过实证分析展示了在合成数据上训练的这一长期存在的诅咒如何在这一特定设置中显现。与基于金标签训练的模型相比，我们不仅观察到预期的准确率和F1分数性能下降，还观察到跨训练运行的稳定性降低以及过早的性能平台期。这些发现对这类方法在现实世界应用中的可靠性提出了质疑。我们通过错误传播的视角对观察到的现象进行了情境化，并提出了几种实用的缓解策略，包括基于熵的过滤和集成技术。尽管这些启发式方法提供了部分缓解，但它们并未完全解决将非随机错误从LLM标注传播到较小分类器的固有风险，强调了在高风险文本分类任务中应用此工作流时需要谨慎。
英文摘要：
Using LLM-generated labels to fine-tune smaller encoder-only models for text classification has gained popularity in various settings. While this approach may be justified in simple and low-stakes applications, we conduct empirical analysis to demonstrate how the perennial curse of training on synthetic data manifests itself in this specific setup. Compared to models trained on gold labels, we observe not only the expected performance degradation in accuracy and F1 score, but also increased instability across training runs and premature performance plateaus. These findings cast doubts on the reliability of such approaches in real-world applications. We contextualize the observed phenomena through the lens of error propagation and offer several practical mitigation strategies, including entropy-based filtering and ensemble techniques. Although these heuristics offer partial relief, they do not fully resolve the inherent risks of propagating non-random errors from LLM annotations to smaller classifiers, underscoring the need for caution when applying this workflow in high-stakes text classification tasks.
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。
