Kharitonov, EugeneBaroni, Marco2022-12-022022-12-022020Kharitonov, E, Baroni, M. Emergent language generalization and acquisition speed are not tied to compositionality. In: Proceedings of the 3rd BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP;2020 Nov 20; [online]. Stroudsburg: Association for Computational Linguistics;2020. p. 11-5. DOI: 10.18653/v1/2020.blackboxnlp-1.2http://hdl.handle.net/10230/55069Comunicació presentada a la 3rd BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP, celebrada el 20 de novembre de 2020 de manera virtual.Studies of discrete languages emerging when neural agents communicate to solve a joint task often look for evidence of compositional structure. This stems for the expectation that such a structure would allow languages to be acquired faster by the agents and enable them to generalize better. We argue that these beneficial properties are only loosely connected to compositionality. In two experiments, we demonstrate that, depending on the task, noncompositional languages might show equal, or better, generalization performance and acquisition speed than compositional ones. Further research in the area should be clearer about what benefits are expected from compositionality, and how the latter would lead to them.application/pdfeng© ACL, Creative Commons Attribution 4.0 LicenseAprenentatge automàticLingüística computacionalinfo:eu-repo/semantics/conferenceObjecthttp://dx.doi.org/10.18653/v1/2020.blackboxnlp-1.2info:eu-repo/semantics/openAccess