001005477 001__ 1005477 001005477 005__ 20230324201755.0 001005477 037__ $$aFZJ-2023-01493 001005477 041__ $$aEnglish 001005477 1001_ $$0P:(DE-Juel1)176538$$aRathkopf, Charles$$b0$$ufzj 001005477 1112_ $$aKimball Union Academy$$conline event$$d2023-01-13 - $$wUSA 001005477 245__ $$aDo Large Language Models Understand Meaning? 001005477 260__ $$c2023 001005477 3367_ $$033$$2EndNote$$aConference Paper 001005477 3367_ $$2DataCite$$aOther 001005477 3367_ $$2BibTeX$$aINPROCEEDINGS 001005477 3367_ $$2ORCID$$aLECTURE_SPEECH 001005477 3367_ $$0PUB:(DE-HGF)31$$2PUB:(DE-HGF)$$aTalk (non-conference)$$btalk$$mtalk$$s1679639876_18831$$xOther 001005477 3367_ $$2DINI$$aOther 001005477 520__ $$aIt is curiously difficult to articulate the capacities of large language modelswithout getting yourself into philosophically controversial terrain. In this talk Iexplain why. The talk has three parts. In the first, I give a sketch of how largelanguage models are built, with particular attention to the way words arerepresented as vector quantities. In the second, I describe the various ways inwhich the capacities of language models have been tested empirically. In thethird, I provide the main philosophical argument. I argue that, in order tounderstand what large language models are, we must reject the seeminglyinnocent metaphysical principle that everything in the world either has a mindor it does not. 001005477 536__ $$0G:(DE-HGF)POF4-5255$$a5255 - Neuroethics and Ethics of Information (POF4-525)$$cPOF4-525$$fPOF IV$$x0 001005477 909CO $$ooai:juser.fz-juelich.de:1005477$$pVDB 001005477 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)176538$$aForschungszentrum Jülich$$b0$$kFZJ 001005477 9131_ $$0G:(DE-HGF)POF4-525$$1G:(DE-HGF)POF4-520$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5255$$aDE-HGF$$bKey Technologies$$lNatural, Artificial and Cognitive Information Processing$$vDecoding Brain Organization and Dysfunction$$x0 001005477 9141_ $$y2023 001005477 9201_ $$0I:(DE-Juel1)INM-7-20090406$$kINM-7$$lGehirn & Verhalten$$x0 001005477 980__ $$atalk 001005477 980__ $$aVDB 001005477 980__ $$aI:(DE-Juel1)INM-7-20090406 001005477 980__ $$aUNRESTRICTED