@inproceedings{67db8ec1756647c4a42e1fb8a2ae3477,

title = {A Benchmark for Recipe Understanding in Artificial Agents},

author = {Jens Nevens and Robin De Haes and Rachel Ringe and Mihai Pomarlan and Robert Porzel and Katrien Beuls and {Van Eecke, Paul}},

url = {https://aclanthology.org/2024.lrec-main.3.pdf},

year  = {2024},

date = {2024-05-01},

urldate = {2024-05-01},

booktitle = {Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation},

pages = {22–42},

abstract = {This paper introduces a novel benchmark that has been designed as a test bed for evaluating whether artificial agents are able to understand how to perform everyday activities, with a focus on the cooking domain. Understanding how to cook recipes is a highly challenging endeavour due to the underspecified and grounded nature of recipe texts, combined with the fact that recipe execution is a knowledge-intensive and precise activity. The benchmark comprises a corpus of recipes, a procedural semantic representation language of cooking actions, qualitative and quantitative kitcVan Van Eeckehen simulators, and a standardised evaluation procedure. Concretely, the benchmark task consists in mapping a recipe formulated in natural language to a set of cooking actions that is precise enough to be executed in the simulated kitchen and yields the desired dish. To overcome the challenges inherent to recipe execution, this mapping process needs to incorporate reasoning over the recipe text, the state of the simulated kitchen environment, common-sense knowledge, knowledge of the cooking domain, and the action space of a virtual or robotic chef. This benchmark thereby addresses the growing interest in human-centric systems that combine natural language processing and situated reasoning to perform everyday activities.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{7de0085126684efea1d1f112044ac15d,

title = {Decentralised Emergence of Robust and Adaptive Linguistic Conventions in Populations of Autonomous Agents Grounded in Continuous Worlds},

author = {Jérôme Botoko Ekila and Jens Nevens and Lara Verheyen and Katrien Beuls and {Van Eecke, Paul}},

url = {https://doi.org/10.48550/arXiv.2401.08461},

year  = {2024},

date = {2024-01-01},

urldate = {2024-01-01},

booktitle = {Proc. of the 23rd International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2024)},

pages = {2168–2170},

publisher = {International Foundation for Autonomous Agents and Multiagent Systems},

abstract = {The field of emergent communication investigates how populations of agents can learn to communicate to solve tasks cooperatively. This paper introduces a methodology for establishing linguistic conventions within populations of autonomous agents in a fully decentralised manner. These conventions enable agents to refer to arbitrary entities in their environment. In our approach, agents participate in reference-based language games, wherein agents aim to direct each other’s attention to specific entities in their shared environment by using linguistic conventions. Importantly, our methodology is directly applicable to any dataset that describes entities using continuously valued features. Through repeated local communicative interactions, agents gradually build up a linguistic inventory, associating words with conceptual representations. We validate our methodology through six experiments on three large tabular datasets, demonstrating its effectiveness in facilitating the emergence of linguistic conventions within populations of autonomous agents. Furthermore, the experiments showcase the robustness of the methodology against sensor defects, its ability to handle noise observations and uncalibrated sensors, suitability for continual learning, and capacity to foster self-adapting languages in response to environmental changes and communicative needs.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{f28b58ab35e340028a223e926809ad7f,

title = {Neuro-Symbolic Procedural Semantics for Reasoning-Intensive Visual Dialogue Tasks},

author = {Lara Verheyen and Jérôme Botoko Ekila and Jens Nevens and {Van Eecke, Paul} and Katrien Beuls},

editor = {Kobi Gal and Ann Nowé and Grzegorz J. Nalepa and Roy Fairstein and Roxana Rădulescu},

url = {https://doi.org/10.3233/FAIA230544},

isbn = {978-1-64368-436-9},

year  = {2023},

date = {2023-09-28},

urldate = {2023-09-28},

booktitle = {ECAI 2023 - 26th European Conference on Artificial Intelligence, including 12th Conference on Prestigious Applications of Intelligent Systems, PAIS 2023 - Proceedings},

pages = {2419 – 2426},

series = {Frontiers in Artificial Intelligence and Applications},

abstract = {This paper introduces a novel approach to visual dialogue that is based on neuro-symbolic procedural semantics. The approach builds further on earlier work on procedural semantics for visual question answering and expands it on the one hand with neuro-symbolic reasoning operations, and on the other hand with mechanisms that handle the challenges that are inherent to dialogue, in particular the incremental nature of the information that is conveyed. Concretely, we introduce (i) the use of a conversation memory as a data structure that explicitly and incrementally represents the information that is expressed during the subsequent turns of a dialogue, and (ii) the design of a neuro-symbolic procedural semantic representation that is grounded in both visual input and the conversation memory. We validate the methodology using the reasoning- intensive MNIST Dialog and CLEVR-Dialog benchmark challenges and achieve a question-level accuracy of 99.8% and 99.2% respectively. The methodology presented in this paper responds to the growing interest in the field of artificial intelligence in solving tasks that involve both low-level perception and high-level reasoning using a combination of neural and symbolic techniques.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{doumen2023modelling,

title = {Modelling Language Acquisition through Syntactico-Semantic Pattern Finding},

author = {Jonas Doumen and Katrien Beuls and {Van Eecke, Paul}},

url = {https://aclanthology.org/2023.findings-eacl.99/},

year  = {2023},

date = {2023-05-01},

urldate = {2023-05-01},

booktitle = {Findings of the Association for Computational Linguistics: EACL 2023},

pages = {1317-1327},

publisher = {Association for Computational Linguistics},

abstract = {Usage-based theories of language acquisition have extensively documented the processes by which children acquire language through communicative interaction. Notably, Tomasello (2003) distinguishes two main cognitive capacities that underlie human language acquisition: intention reading and pattern finding. Intention reading is the process by which children try to continuously reconstruct the intended meaning of their interlocutors. Pattern finding refers to the process that allows them to distil linguistic schemata from multiple communicative interactions. Even though the fields of cognitive science and psycholinguistics have studied these processes in depth, no faithful computational operationalisations of these mechanisms through which children learn language exist to date. The research on which we report in this paper aims to fill part of this void by introducing a computational operationalisation of syntactico-semantic pattern finding. Concretely, we present a methodology for learning grammars based on similarities and differences in the form and meaning of linguistic observations alone. Our methodology is able to learn compositional lexical and item-based constructions of variable extent and degree of abstraction, along with a network of emergent syntactic categories. We evaluate our methodology on the CLEVR benchmark dataset and show that the methodology allows for fast, incremental and effective learning. The constructions and categorial network that result from the learning process are fully transparent and bidirectional, facilitating both language comprehension and production. Theoretically, our model provides computational evidence for the learnability of usage-based constructionist theories of language acquisition. Practically, the techniques that we present facilitate the learning of computationally tractable, usage-based construction grammars, which are applicable for natural language understanding and production tasks.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{beuls2023fluid,

title = {Fluid Construction Grammar: State of the Art and Future Outlook},

author = {Katrien Beuls and {Van Eecke, Paul}},

url = {https://aclanthology.org/2023.cxgsnlp-1.6/},

year  = {2023},

date = {2023-03-01},

urldate = {2023-03-01},

booktitle = {Proceedings of the First International Workshop on Construction Grammars and NLP (CxGs+NLP, GURT/SyntaxFest 2023)},

pages = {41-50},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{9cbc6213e4524fa4b321a7aec9a7a65a,

title = {The Candide model: How narratives emerge where observations meet beliefs},

author = {{Van Eecke, Paul} and Lara Verheyen and Tom Willaert and Katrien Beuls},

editor = {Nader Akoury and Elizabeth Clark and Mohit Iyyer and Snigdha Chaturvedi and Faeze Brahman and Khyathi Raghavi Chandu},

url = {https://aclanthology.org/2023.wnu-1.7/},

year  = {2023},

date = {2023-01-01},

urldate = {2023-01-01},

booktitle = {5th Workshop on Narrative Understanding, WNU 2023 - Proceedings of the Workshop},

pages = {48–57},

publisher = {Association for Computational Linguistics},

series = {Proceedings of the Annual Meeting of the Association for Computational Linguistics},

abstract = {This paper presents the Candide model as a computational architecture for modelling human-like, narrative-based language understanding. The model starts from the idea that narratives emerge through the process of interpreting novel linguistic observations, such as utterances, paragraphs and texts, with respect to previously acquired knowledge and beliefs. Narratives are personal, as they are rooted in past experiences, and constitute perspectives on the world that might motivate different interpretations of the same observations. Concretely, the Candide model operationalises this idea by dynamically modelling the belief systems and background knowledge of individual agents, updating these as new linguistic observations come in, and exposing them to a logic reasoning engine that reveals the possible sources of divergent interpretations. Apart from introducing the foundational ideas, we also present a proof-of-concept implementation that demonstrates the approach through a number of illustrative examples.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{nevens2022language,

title = {Language Acquisition through Intention Reading and Pattern Finding},

author = {Jens Nevens and Jonas Doumen and {Van Eecke, Paul} and Katrien Beuls},

url = {https://aclanthology.org/2022.coling-1.2/},

year  = {2022},

date = {2022-10-01},

urldate = {2022-10-01},

booktitle = {Proceedings of the 29th International Conference on Computational Linguistics},

pages = {15-25},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}