什么的大型模型?将工程成就误认为人类的语言能力

IF 1.7 2区 文学 Q2 EDUCATION & EDUCATIONAL RESEARCH Language Sciences Pub Date : 2024-08-21 DOI:10.1016/j.langsci.2024.101672
{"title":"什么的大型模型?将工程成就误认为人类的语言能力","authors":"","doi":"10.1016/j.langsci.2024.101672","DOIUrl":null,"url":null,"abstract":"<div><p>In this paper we argue that key, often sensational and misleading, claims regarding linguistic capabilities of Large Language Models (LLMs) are based on at least two unfounded assumptions: the <em>assumption of language completeness</em> and the <em>assumption of data completeness</em>. Language completeness assumes that a distinct and complete thing such as “a natural language” exists, the essential characteristics of which can be effectively and comprehensively modelled by an LLM. The assumption of data completeness relies on the belief that a language can be quantified and wholly captured by data. Work within the enactive approach to cognitive science makes clear that, rather than a distinct and complete thing, language is a means or way of acting. Languaging is not the kind of thing that can admit of a complete or comprehensive modelling. From an enactive perspective we identify three key characteristics of enacted language; <em>embodiment</em>, <em>participation</em>, and <em>precariousness</em>, that are absent in LLMs, and likely incompatible in principle with current architectures. We argue that these absences imply that LLMs are not now and cannot in their present form be linguistic agents the way humans are. We illustrate the point in particular through the phenomenon of “algospeak”, a recently described pattern of high-stakes human language activity in heavily controlled online environments. On the basis of these points, we conclude that sensational and misleading claims about LLM agency and capabilities emerge from a deep misconception of both what human language is and what LLMs are.</p></div>","PeriodicalId":51592,"journal":{"name":"Language Sciences","volume":null,"pages":null},"PeriodicalIF":1.7000,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S0388000124000615/pdfft?md5=f7f3281a359df35af751aca63248e4e7&pid=1-s2.0-S0388000124000615-main.pdf","citationCount":"0","resultStr":"{\"title\":\"Large models of what? Mistaking engineering achievements for human linguistic agency\",\"authors\":\"\",\"doi\":\"10.1016/j.langsci.2024.101672\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>In this paper we argue that key, often sensational and misleading, claims regarding linguistic capabilities of Large Language Models (LLMs) are based on at least two unfounded assumptions: the <em>assumption of language completeness</em> and the <em>assumption of data completeness</em>. Language completeness assumes that a distinct and complete thing such as “a natural language” exists, the essential characteristics of which can be effectively and comprehensively modelled by an LLM. The assumption of data completeness relies on the belief that a language can be quantified and wholly captured by data. Work within the enactive approach to cognitive science makes clear that, rather than a distinct and complete thing, language is a means or way of acting. Languaging is not the kind of thing that can admit of a complete or comprehensive modelling. From an enactive perspective we identify three key characteristics of enacted language; <em>embodiment</em>, <em>participation</em>, and <em>precariousness</em>, that are absent in LLMs, and likely incompatible in principle with current architectures. We argue that these absences imply that LLMs are not now and cannot in their present form be linguistic agents the way humans are. We illustrate the point in particular through the phenomenon of “algospeak”, a recently described pattern of high-stakes human language activity in heavily controlled online environments. On the basis of these points, we conclude that sensational and misleading claims about LLM agency and capabilities emerge from a deep misconception of both what human language is and what LLMs are.</p></div>\",\"PeriodicalId\":51592,\"journal\":{\"name\":\"Language Sciences\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":1.7000,\"publicationDate\":\"2024-08-21\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.sciencedirect.com/science/article/pii/S0388000124000615/pdfft?md5=f7f3281a359df35af751aca63248e4e7&pid=1-s2.0-S0388000124000615-main.pdf\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Language Sciences\",\"FirstCategoryId\":\"98\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0388000124000615\",\"RegionNum\":2,\"RegionCategory\":\"文学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"EDUCATION & EDUCATIONAL RESEARCH\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Language Sciences","FirstCategoryId":"98","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0388000124000615","RegionNum":2,"RegionCategory":"文学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"EDUCATION & EDUCATIONAL RESEARCH","Score":null,"Total":0}
引用次数: 0

摘要

在本文中,我们认为有关大型语言模型(LLMs)语言能力的关键性、往往是耸人听闻和误导性的说法至少基于两个毫无根据的假设:语言完整性假设和数据完整性假设。语言完整性假定存在一个独特而完整的事物,如 "自然语言",其基本特征可由 LLM 有效而全面地建模。数据完备性的假设则是基于这样一种信念,即语言可以量化并完全由数据捕捉。认知科学的能动方法清楚地表明,与其说语言是一种独特而完整的事物,不如说它是一种行动的手段或方式。语言不是一种可以完整或全面建模的事物。从能动的角度出发,我们确定了被应用语言的三个关键特征:体现、参与和不稳定性,这些特征在语言学习工具中都不存在,而且很可能与当前的架构原则上不兼容。我们认为,这些特征的缺失意味着 LLMs 现在不是、也不可能以其现有形式成为人类那样的语言代理。我们特别通过 "algospeak "现象来说明这一点,"algospeak "是最近描述的在严格控制的在线环境中人类高风险语言活动的一种模式。基于以上观点,我们得出结论:关于 LLM 的作用和能力的耸人听闻和误导性的说法,源于对人类语言和 LLM 的深刻误解。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Large models of what? Mistaking engineering achievements for human linguistic agency

In this paper we argue that key, often sensational and misleading, claims regarding linguistic capabilities of Large Language Models (LLMs) are based on at least two unfounded assumptions: the assumption of language completeness and the assumption of data completeness. Language completeness assumes that a distinct and complete thing such as “a natural language” exists, the essential characteristics of which can be effectively and comprehensively modelled by an LLM. The assumption of data completeness relies on the belief that a language can be quantified and wholly captured by data. Work within the enactive approach to cognitive science makes clear that, rather than a distinct and complete thing, language is a means or way of acting. Languaging is not the kind of thing that can admit of a complete or comprehensive modelling. From an enactive perspective we identify three key characteristics of enacted language; embodiment, participation, and precariousness, that are absent in LLMs, and likely incompatible in principle with current architectures. We argue that these absences imply that LLMs are not now and cannot in their present form be linguistic agents the way humans are. We illustrate the point in particular through the phenomenon of “algospeak”, a recently described pattern of high-stakes human language activity in heavily controlled online environments. On the basis of these points, we conclude that sensational and misleading claims about LLM agency and capabilities emerge from a deep misconception of both what human language is and what LLMs are.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Language Sciences
Language Sciences Multiple-
CiteScore
2.90
自引率
0.00%
发文量
38
期刊介绍: Language Sciences is a forum for debate, conducted so as to be of interest to the widest possible audience, on conceptual and theoretical issues in the various branches of general linguistics. The journal is also concerned with bringing to linguists attention current thinking about language within disciplines other than linguistics itself; relevant contributions from anthropologists, philosophers, psychologists and sociologists, among others, will be warmly received. In addition, the Editor is particularly keen to encourage the submission of essays on topics in the history and philosophy of language studies, and review articles discussing the import of significant recent works on language and linguistics.
期刊最新文献
A study of visual path expressions in Mandarin Chinese from the perspective of motion event typology The etymology of opaque place names based on a cognitive and interdisciplinary method Third-way linguistics: generative and usage-based theories are both right Further semantic change of the derogatory sociomorpheme tái in Chinese gender-related Internet neologisms Science mapping the literature in Applied Linguistics secondary research: navigating knowledge evolution from an epistemic perspective (1970–2022)
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1