2024 |
Liesbet De Vos Jens Nevens, Paul Van Eecke ; Beuls, Katrien Construction grammar and procedural semantics for human-interpretable grounded language processing Journal Article Linguistics Vanguard, (0), 2024. Abstract | BibTeX | Tags: Computational Construction Grammar, Fluid Construction Grammar, grounded language understanding, procedural semantics @article{de2024construction, title = {Construction grammar and procedural semantics for human-interpretable grounded language processing}, author = {Liesbet De Vos, Jens Nevens, Paul Van Eecke and Katrien Beuls}, year = {2024}, date = {2024-03-15}, journal = {Linguistics Vanguard}, number = {0}, abstract = {Groundedlanguageprocessingisacrucialcomponentinmanyartificialintelligencesystems,asitallows agents to communicate about their physical surroundings. State-of-the-art approaches typically employ deep learning techniques that perform end-to-end mappings between natural language expressions and representations grounded in the environment. Although these techniques achieve high levels of accuracy, they are often criticized for their lack of interpretability and their reliance on large amounts of training data. As an alternative, we propose a fully interpretable, data-efficient architecture for grounded language processing. The architecture is based on two main components. The first component comprises an inventory of human-interpretable concepts learned through task-based communicative interactions. These concepts connect the sensorimotor experiences of an agent to meaningful symbols that can be used for reasoning operations. The second component is a computational con- struction grammar that maps between natural language expressions and procedural semantic representations. These representations are grounded through their integration with the learned concepts. We validate the archi- tecture using a variation on the CLEVR benchmark, achieving an accuracy of 96 %. Our experiments demonstrate that the integration of a computational construction grammar with an inventory of interpretable grounded con- cepts can effectively achieve human-interpretable grounded language processing in the CLEVR environment.}, keywords = {Computational Construction Grammar, Fluid Construction Grammar, grounded language understanding, procedural semantics}, pubstate = {published}, tppubtype = {article} } Groundedlanguageprocessingisacrucialcomponentinmanyartificialintelligencesystems,asitallows agents to communicate about their physical surroundings. State-of-the-art approaches typically employ deep learning techniques that perform end-to-end mappings between natural language expressions and representations grounded in the environment. Although these techniques achieve high levels of accuracy, they are often criticized for their lack of interpretability and their reliance on large amounts of training data. As an alternative, we propose a fully interpretable, data-efficient architecture for grounded language processing. The architecture is based on two main components. The first component comprises an inventory of human-interpretable concepts learned through task-based communicative interactions. These concepts connect the sensorimotor experiences of an agent to meaningful symbols that can be used for reasoning operations. The second component is a computational con- struction grammar that maps between natural language expressions and procedural semantic representations. These representations are grounded through their integration with the learned concepts. We validate the archi- tecture using a variation on the CLEVR benchmark, achieving an accuracy of 96 %. Our experiments demonstrate that the integration of a computational construction grammar with an inventory of interpretable grounded con- cepts can effectively achieve human-interpretable grounded language processing in the CLEVR environment. |
2022 |
van Trijp, Remi ; Beuls, Katrien; Van Eecke, Paul The FCG Editor: An innovative environment for engineering computational construction grammars Journal Article PLOS ONE, 17 (6), pp. 1-27, 2022. Abstract | Links | BibTeX | Tags: Computational Construction Grammar, Fluid Construction Grammar @article{vantrijp2022fcg, title = {The FCG Editor: An innovative environment for engineering computational construction grammars}, author = {Remi {van Trijp} and Katrien Beuls and Paul {Van Eecke}}, doi = {10.1371/journal.pone.0269708}, year = {2022}, date = {2022-01-01}, journal = {PLOS ONE}, volume = {17}, number = {6}, pages = {1-27}, publisher = {Public Library of Science}, abstract = {Since its inception in the mid-eighties, the field of construction grammar has been steadily growing and constructionist approaches to language have by now become a mainstream paradigm for linguistic research. While the construction grammar community has traditionally focused on theoretical, experimental and corpus-based research, the importance of computational methodologies is now rapidly increasing. This movement has led to the establishment of a number of exploratory computational construction grammar formalisms, which facilitate the implementation of construction grammars, as well as their use for language processing purposes. Yet, implementing large grammars using these formalisms still remains a challenging task, partly due to a lack of powerful and user-friendly tools for computational construction grammar engineering. In order to overcome this obstacle, this paper introduces the FCG Editor, a dedicated and innovative integrated development environment for the Fluid Construction Grammar formalism. Offering a straightforward installation and a user-friendly, interactive interface, the FCG Editor is an accessible, yet powerful tool for construction grammarians who wish to operationalise their construction grammar insights and analyses in order to computationally verify them, corroborate them with corpus data, or integrate them in language technology applications.}, keywords = {Computational Construction Grammar, Fluid Construction Grammar}, pubstate = {published}, tppubtype = {article} } Since its inception in the mid-eighties, the field of construction grammar has been steadily growing and constructionist approaches to language have by now become a mainstream paradigm for linguistic research. While the construction grammar community has traditionally focused on theoretical, experimental and corpus-based research, the importance of computational methodologies is now rapidly increasing. This movement has led to the establishment of a number of exploratory computational construction grammar formalisms, which facilitate the implementation of construction grammars, as well as their use for language processing purposes. Yet, implementing large grammars using these formalisms still remains a challenging task, partly due to a lack of powerful and user-friendly tools for computational construction grammar engineering. In order to overcome this obstacle, this paper introduces the FCG Editor, a dedicated and innovative integrated development environment for the Fluid Construction Grammar formalism. Offering a straightforward installation and a user-friendly, interactive interface, the FCG Editor is an accessible, yet powerful tool for construction grammarians who wish to operationalise their construction grammar insights and analyses in order to computationally verify them, corroborate them with corpus data, or integrate them in language technology applications. |
2021 |
Beuls, Katrien; Van Eecke, Paul ; Cangalovic, Vanja Sophie A Computational Construction Grammar Approach to Semantic Frame Extraction Journal Article Linguistics Vanguard, 7 (1), pp. 20180015, 2021. Links | BibTeX | Tags: Computational Construction Grammar, Fluid Construction Grammar @article{beuls2021computational, title = {A Computational Construction Grammar Approach to Semantic Frame Extraction}, author = {Katrien Beuls and Paul {Van Eecke} and Vanja Sophie Cangalovic}, url = {https://ehai.ai.vub.ac.be/assets/pdfs/beuls2021computational.pdf}, doi = {https://doi.org/10.1515/lingvan-2018-0015}, year = {2021}, date = {2021-01-01}, journal = {Linguistics Vanguard}, volume = {7}, number = {1}, pages = {20180015}, publisher = {De Gruyter}, keywords = {Computational Construction Grammar, Fluid Construction Grammar}, pubstate = {published}, tppubtype = {article} } |
2018 |
Van Eecke, Paul ; Beuls, Katrien Exploring the Creative Potential of Computational Construction Grammar Journal Article Zeitschrift für Anglistik und Amerikanistik, 66 (3), pp. 341–355, 2018, ISSN: 2196-4726. Abstract | Links | BibTeX | Tags: Computational Construction Grammar, Creativity, Fluid Construction Grammar @article{30da883aaecd4f11814c93a15a89855f, title = {Exploring the Creative Potential of Computational Construction Grammar}, author = {Paul {Van Eecke} and Katrien Beuls}, url = {https://ehai.ai.vub.ac.be/assets/pdfs/fcg-creativity.pdf}, doi = {10.1515/zaa-2018-0029}, issn = {2196-4726}, year = {2018}, date = {2018-01-01}, journal = {Zeitschrift für Anglistik und Amerikanistik}, volume = {66}, number = {3}, pages = {341--355}, publisher = {Walter de Gruyter GmbH}, abstract = {Computational construction grammar aims to provide concrete processing models that operationalise construction grammar accounts of the different aspects of language. This paper discusses the computational mechanisms that allow construction grammar models to exhibit, to a certain extent, the creativity and inventiveness that is observed in human language use. It addresses two main types of language-related creativity. The first type concerns the ‘free combination of constructions,’ which gives rise to the open-endedness of language. The second type concerns the ‘appropriate violation of usual constraints’ that permits language users to go beyond what is possible when adhering to the usual constraints of the language, and be truly creative by relaxing these constraints and by introducing novel constructions. All mechanisms and examples discussed in this paper are fully operationalised and implemented in Fluid Construction Grammar.}, keywords = {Computational Construction Grammar, Creativity, Fluid Construction Grammar}, pubstate = {published}, tppubtype = {article} } Computational construction grammar aims to provide concrete processing models that operationalise construction grammar accounts of the different aspects of language. This paper discusses the computational mechanisms that allow construction grammar models to exhibit, to a certain extent, the creativity and inventiveness that is observed in human language use. It addresses two main types of language-related creativity. The first type concerns the ‘free combination of constructions,’ which gives rise to the open-endedness of language. The second type concerns the ‘appropriate violation of usual constraints’ that permits language users to go beyond what is possible when adhering to the usual constraints of the language, and be truly creative by relaxing these constraints and by introducing novel constructions. All mechanisms and examples discussed in this paper are fully operationalised and implemented in Fluid Construction Grammar. |
Publications
2024 |
Construction grammar and procedural semantics for human-interpretable grounded language processing Journal Article Linguistics Vanguard, (0), 2024. |
2022 |
The FCG Editor: An innovative environment for engineering computational construction grammars Journal Article PLOS ONE, 17 (6), pp. 1-27, 2022. |
2021 |
A Computational Construction Grammar Approach to Semantic Frame Extraction Journal Article Linguistics Vanguard, 7 (1), pp. 20180015, 2021. |
2018 |
Exploring the Creative Potential of Computational Construction Grammar Journal Article Zeitschrift für Anglistik und Amerikanistik, 66 (3), pp. 341–355, 2018, ISSN: 2196-4726. |