高效符号通信编码的演化
Evolution of Efficient Symbolic Communication Codes
AI总结：
该研究将语言结构视为通信编码演化产物，通过最大化反熵等指标开展元学习以优化分词，经多语言实验验证了不同指标对各语言及子词解析的差异化作用。
摘要：
本文探讨了人类自然语言结构如何可被视为人际通信编码演化的产物，其演化目标是最大化反熵、压缩因子和跨拆分F1分数这类跨文化、跨语言的指标。这项探索是更大规模无监督语言学习研究的一部分，研究尝试通过最大化上述指标，在超参数空间中开展元学习，以基于“真值”语言结构最大化F1分数。本文呈现了针对俄语、汉语和英语的跨语言词级分词tokenization研究的初步结果，以及针对英语的子词分割或称形态解析研究的初步结果。研究发现，词级分词或称tokenization形成的语言结构受所有这些指标的驱动：反熵与英语、俄语的相关性更强，而压缩因子对汉语而言更具特异性。针对英语词汇的子词分割或称形态解析研究显示，已发现的压缩效应与压缩因子直接相关，而令人意外的是，其与反熵的关联则呈反向关系。
英文摘要：
The paper explores how the human natural language structure can be seen as a product of evolution of inter-personal communication code, targeting maximisation of such culture-agnostic and cross-lingual metrics such as anti-entropy, compression factor and cross-split F1 score. The exploration is done as part of a larger unsupervised language learning effort, the attempt is made to perform meta-learning in a space of hyper-parameters maximising F1 score based on the "ground truth" language structure, by means of maximising the metrics mentioned above. The paper presents preliminary results of cross-lingual word-level segmentation tokenisation study for Russian, Chinese and English as well as subword segmentation or morphological parsing study for English. It is found that language structure form the word-level segmentation or tokenisation can be found as driven by all of these metrics, anti-entropy being more relevant to English and Russian while compression factor more specific for Chinese. The study for subword segmentation or morphological parsing on English lexicon has revealed straight connection between the compression been found to be associated with compression factor, while, surprising, the same connection with anti-entropy has turned to be the inverse.
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。
