arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2102.08424cs.CL
收藏

神经形态学拐点中搜索错误的搜索

Searching for Search Errors in Neural Morphological Inflection

ETH Zürich(苏黎世联邦理工学院);University of Cambridge(剑桥大学)

Martina Forster, Clara Meister, Ryan Cotterell

摘要:

神经序列到序列模型是当前语言生成任务的主要选择。然而,在单词级别的任务中,这些模型的精确推理表明,空字符串通常是全局最优的。先前的工作推测这种现象是由于语言生成的神经模型的不足。然而,在形态拐点的情况下,我们发现,空字符串几乎从来都不是模型下的最可能的解决方案。此外,贪婪搜索通常会找到全局最优值。这些观察结果表明,许多神经模型的校准不良可能源于特定任务子集的特征,而不是这些模型对语言生成的普遍不适。

英文摘要:

Neural sequence-to-sequence models are currently the predominant choice for language generation tasks. Yet, on word-level tasks, exact inference of these models reveals the empty string is often the global optimum. Prior works have speculated this phenomenon is a result of the inadequacy of neural models for language generation. However, in the case of morphological inflection, we find that the empty string is almost never the most probable solution under the model. Further, greedy search often finds the global optimum. These observations suggest that the poor calibration of many neural models may stem from characteristics of a specific subset of tasks rather than general ill-suitedness of such models for language generation.

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。

↑