| 1. |
- Adouane, Wafia, 1985, et al.
(författare)
-
A Comparison of Character Neural Language Model and Bootstrapping for Language Identification in Multilingual Noisy Texts
- 2018
-
Ingår i: Proceedings of the Second Workshop on Subword and Character Level Models in NLP (SCLeM), June 6, 2018 New Orleans, Louisiana. - New Orleans, Louisiana USA. - 9781948087186
-
Konferensbidrag (refereegranskat)abstract
- This paper seeks to examine the effect of including background knowledge in the form of character pre-trained neural language model (LM), and data bootstrapping to overcome the problem of unbalanced limited resources. As a test, we explore the task of language identification in mixed-language short non-edited texts with an under-resourced language, namely the case of Algerian Arabic for which both labelled and unlabelled data are limited. We compare the performance of two traditional machine learning methods and a deep neural networks (DNNs) model. The results show that overall DNNs perform better on labelled data for the majority categories and struggle with the minority ones. While the effect of the untokenised and unlabelled data encoded as LM differs for each category, bootstrapping, however, improves the performance of all systems and all categories. These methods are language independent and could be generalised to other under-resourced languages for which a small labelled data and a larger unlabelled data are available.
|
|
| 2. |
- Adouane, Wafia, 1985, et al.
(författare)
-
Identification of Languages in Algerian Arabic Multilingual Documents
- 2017
-
Ingår i: Proceedings of The Third Arabic Natural Language Processing Workshop (WANLP), Valencia, Spain, April 3, 2017. - Valencia, Spain : Association for Computational Linguistics. - 9781945626449
-
Konferensbidrag (refereegranskat)abstract
- This paper presents a language identification system designed to detect the language of each word, in its context, in a multilingual documents as generated in social media by bilingual/multilingual communities, in our case speakers of Algerian Arabic. We frame the task as a sequence tagging problem and use supervised machine learning with standard methods like HMM and Ngram classifi- cation tagging. We also experiment with a lexicon-based method. Combining all the methods in a fall-back mechanism and introducing some linguistic rules, to deal with unseen tokens and ambiguous words, gives an overall accuracy of 93.14%. Finally, we introduced rules for language identification from sequences of recognised words.
|
|
| 3. |
- Adouane, Wafia, 1985, et al.
(författare)
-
Improving Neural Network Performance by Injecting Background Knowledge: Detecting Code-switching and Borrowing in Algerian texts
- 2018
-
Ingår i: Proceedings of the Third Workshop on Computational Approaches to Linguistic Code-switching, Melbourne, Australia, July 19, 2018. - Melbourne, Australia : Association for Computational Linguistics. - 9781948087452
-
Konferensbidrag (refereegranskat)abstract
- We explore the effect of injecting back- ground knowledge to different deep neural network (DNN) configurations in order to mitigate the problem of the scarcity of annotated data when applying these models on datasets of low-resourced languages. The background knowledge is encoded in the form of lexicons and pre-trained sub-word embeddings. The DNN models are evaluated on the task of detecting code-switching and borrowing points in non-standardised user-generated Algerian texts. Overall results show that DNNs benefit from adding background knowledge. However, the gain varies between models and categories. The proposed DNN architectures are generic and could be applied to other low-resourced languages.
|
|
| 4. |
- Adouane, Wafia, 1985, et al.
(författare)
-
Neural Models for Detecting Binary Semantic Textual Similarity for Algerian and MSA
- 2019
-
Ingår i: Proceedings of the Fourth Arabic Natural Language Processing Workshop, WANLP 2019, Jul 28-Aug 2, Florence, Italy. pp. 78-87. - Florence, Italy : Association for Computational Linguistics. - 9781950737321
-
Konferensbidrag (refereegranskat)abstract
- We explore the extent to which neural networks can learn to identify semantically equivalent sentences from a small variable dataset using an end-to-end training. We collect a new noisy non-standardised user-generated Algerian (ALG) dataset and also translate it to Modern Standard Arabic (MSA) which serves as its regularised counterpart. We compare the performance of various models on both datasets and report the best performing configurations. The results show that relatively simple models composed of 2 LSTM layers outperform by far other more sophisticated attention-based architectures, for both ALG and MSA datasets.
|
|
| 5. |
- Adouane, Wafia, 1985, et al.
(författare)
-
Normalising Non-standardised Orthography in Algerian Code-switched User-generated Data
- 2019
-
Ingår i: The 5th Workshop on Noisy User-generated Text (W-NUT), November 4, 2019, Hong Kong / Wei Xu, Alan Ritter, Tim Baldwin, Afshin Rahimi (Editors). - Stroudsburg, PA : Association for Computational Linguistics. - 9781950737840
-
Konferensbidrag (refereegranskat)abstract
- We work with Algerian, an under-resourced non-standardised Arabic variety, for which we compile a new parallel corpus consist- ing of user-generated textual data matched with normalised and corrected human annota- tions following data-driven and our linguisti- cally motivated standard. We use an end-to- end deep neural model designed to deal with context-dependent spelling correction and nor- malisation. Results indicate that a model with two CNN sub-network encoders and an LSTM decoder performs the best, and that word context matters. Additionally, pre- processing data token-by-token with an edit- distance based aligner significantly improves the performance. We get promising results for the spelling correction and normalisation, as a pre-processing step for downstream tasks, on detecting binary Semantic Textual Similarity.
|
|
| 6. |
- Amanaki, Erini, et al.
(författare)
-
Fine-grained Entailment: Resources for Greek NLI and Precise Entailment
- 2022
-
Ingår i: Proceedings of the Workshop on Dataset Creation for Lower-Resourced Languages within the 13th Language Resources and Evaluation Conference. - Marseille, France : European Language Resources Association (ELRA). - 9782493814067
-
Konferensbidrag (refereegranskat)abstract
- In this paper, we present a number of fine-grained resources for Natural Language Inference (NLI). In particular, we present a number of resources and validation methods for Greek NLI and a resource for precise NLI. First, we extend the Greek version of the FraCaS test suite to include examples where the inference is directly linked to the syntactic/morphological properties of Greek. The new resource contains an additional 428 examples, making it in total a dataset of 774 examples. Expert annotators have been used in order to create the additional resource, while extensive validation of the original Greek version of the FraCaS by non-expert and expert subjects is performed. Next, we continue the work initiated by (CITATION), according to which a subset of the RTE problems have been labeled for missing hypotheses and we present a dataset an order of magnitude larger, annotating the whole SuperGlUE/RTE dataset with missing hypotheses. Lastly, we provide a de-dropped version of the Greek XNLI dataset, where the pronouns that are missing due to the pro-drop nature of the language are inserted. We then run some models to see the effect of that insertion and report the results.
|
|
| 7. |
- Bizzoni, Yuri, 1989, et al.
(författare)
-
Distributional semantic models for detection of textual entailment
- 2016
-
Ingår i: The Sixth Swedish language technology conference (SLTC), Umeå University, 17-18 November, 2016, (ed. Björklund, Johanna and Stymne, Sara). - Umeå : Umeå University.
-
Konferensbidrag (refereegranskat)abstract
- We present our experiments on integrating and evaluating distributional semantics with the recognising textual entailment task (RTE). We consider entailment as semantic similarity between text and hypothesis coupled with additional heuristic, which can be either selecting the top scoring hypothesis or a pre-defined threshold. We show that a distributional model is particularly good at detecting entailment related to “world knowledge”, and that aligning the hypothesis with the text improves detection of lexical entailment.
|
|
| 8. |
- Bizzoni, Yuri, 1989, et al.
(författare)
-
Sky + fire = sunset. Exploring parallels between visually grounded metaphors and image classifiers
- 2020
-
Ingår i: Proceedings of the Second Workshop on Figurative Language Processing (FLP) at ACL-2020, July 9, 2020 / Beata Beigman Klebanov, Ekaterina Shutova, Patricia Lichtenstein, Smaranda Muresan, Chee Wee, Anna Feldman, Debanjan Ghosh (Editors). - Stroudsburg, PA, USA : Association for Computational Linguistics. - 9781952148125
-
Konferensbidrag (refereegranskat)abstract
- This work explores the differences and similarities between neural image classifiers' mis-categorisations and visually grounded metaphors - that we could conceive as intentional mis-categorisations. We discuss the possibility of using automatic image classifiers to approximate human metaphoric behaviours, and the limitations of such frame. We report two pilot experiments to study grounded metaphoricity. In the first we represent metaphors as a form of visual mis-categorisation. In the second we model metaphors as a more flexible, compositional operation in a continuous visual space generated from automatic classification systems.
|
|
| 9. |
- Cano Santín, José Miguel, 1990, et al.
(författare)
-
Fast visual grounding in interaction: bringing few-shot learning with neural networks to an interactive robot
- 2020
-
Ingår i: Proceedings of Conference on Probability and Meaning (PaM-2020), Gothenburg, Sweden (online) / Christine Howes, Stergios Chatzikyriakidis, Adam Ek and Vidya Somashekarappa (eds.). - : Association for Computational Linguistics (ACL). - 2002-9764.
-
Konferensbidrag (refereegranskat)abstract
- The major shortcomings of using neural networks with situated agents are that in incremental interaction very few learning examples are available and that their visual sensory representations are quite different from image caption datasets. In this work we adapt and evaluate a few-shot learning approach, Matching Networks (Vinyals et al., 2016), to conversational strategies of a robot interacting with a human tutor in order to efficiently learn to categorise objects that are presented to it and also investigate to what degree transfer learning from pre-trained models on images from different contexts can improve its performance. We discuss the implications of such learning on the nature of semantic representations the system has learned.
|
|
| 10. |
- Cano Santín, José Miguel, 1990, et al.
(författare)
-
Interactive visual grounding with neural networks
- 2019
-
Ingår i: Proceedings of LondonLogue - Semdial 2019: The 23rd Workshop on the Semantics and Pragmatics of Dialogue, London, 4-6 September 2019. - London, UK : Queen Mary University of London. - 2308-2275.
-
Konferensbidrag (refereegranskat)abstract
- Training strategies for neural networks are not suitable for real time human-robot interaction. Few-shot learning approaches have been developed for low resource scenarios but without the usual teacher/learner supervision. In this work we present a combination of both: a situated dialogue system to teach object names to a robot from its camera images using Matching Networks (Vinyals et al., 2016). We compare the performance of the system with transferred learning from pre-trained models and different conversational strategies with a human tutor.
|
|
| 11. |
|
|
| 12. |
|
|
| 13. |
- Cooper, Robin, 1947, et al.
(författare)
-
A Probabilistic Rich Type Theory for Semantic Interpretation
- 2014
-
Ingår i: Proceedings of the Workshop on Type Theory and Natural Language Semantics (TTNLS),14th Conference of the European Chapter of the Association for Computational Linguistics, Gothenburg, Sweden 2014. ed. by Cooper, Robin, Simon Dobnik, Shalom Lappin and Staffan Larsson. - 9781937284749
-
Konferensbidrag (refereegranskat)abstract
- We propose a probabilistic type theory in which a situation s is judged to be of a type T with probability p. In addition to basic and functional types it includes, inter alia, record types and a notion of typing based on them. The type system is intensional in that types of situations are not reduced to sets of situations. We specify the fragment of a compositional semantics in which truth conditions are replaced by probability conditions. The type system is the interface between classifying situations in perception and computing the semantic interpretations of phrases in natural language.
|
|
| 14. |
- Cooper, Robin, 1947, et al.
(författare)
-
Probabilistic Type Theory and Natural Language Semantics
- 2015
-
Ingår i: Linguistic Issues in Language Technology. - 1945-3590 .- 1945-3604. ; 10:1, s. 1-43
-
Tidskriftsartikel (refereegranskat)abstract
- Type theory has played an important role in specifying the formal con- nection between syntactic structure and semantic interpretation within the history of formal semantics. In recent years rich type theories de- veloped for the semantics of programming languages have become in- fluential in the semantics of natural language. The use of probabilistic reasoning to model human learning and cognition has become an in- creasingly important part of cognitive science. In this paper we offer a probabilistic formulation of a rich type theory, Type Theory with Records (TTR), and we illustrate how this framework can be used to approach the problem of semantic learning. Our probabilistic version of TTR is intended to provide an interface between the cognitive process of classifying situations according to the types that they instantiate, and the compositional semantics of natural language
|
|
| 15. |
- Cooper, Robin, 1947, et al.
(författare)
-
Testing the FraCaS test suite
- 2016
-
Ingår i: Logic and Engineering of Natural Language Semantics 13 (LENLS 13), November 13th, 2016, Tokyo, Japan. - Tachikawa City, Tokyo, Japan : National Institute for Japanese Language and Linguistics. - 9784915905780
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- In this talk I will present some of the background to the project FraCaS which led to the FraCaS test suite. I will also talk about the MultiFraCaS project and some of our ideas for extending this in the future. The examples in the origi- nal test suite were created by semanticists. I will discuss a number of ways in which one could go about verifying these examples. In particular, I will present some preliminary work we have been doing using web-based forms to collect judgements via crowd-sourcing. We will discuss the implications of some prelim- inary results that we have obtained, in particular the possibility of developing a probabilistic semantics. Crowd-sourcing allows us to extend the notion of inference from strict logical inference to inference that is gradient and is prevalent in lexical meaning. The probabilities obtained through crowd-sourcing tell us the likelihood of a native speaker to make a particular conclusion. Eventually, we hope to extend the entire FraCas suite this way. The original aim of the test suite was to provide a way of evaluating compu- tational systems that perform natural language inference. I will talk about some work applying type theory to this task, with as yet partial coverage of the test suite. The crowd-sourcing methods we have been using to evaluate the test suite can also be used to give an empirical basis to predictions made by semantic theories that are difficult to ascertain by only relying on intuitions of a single linguist. I will present some work we have been doing on the semantics of verbal restructuring, a phenomenon whose semantics has been debated and disagreed upon in the literature, and discuss the preliminary results we have.
|
|
| 16. |
- Dannélls, Dana, 1976, et al.
(författare)
-
Beyond strings of characters: Resources meet NLP – Again
- 2022
-
Ingår i: Live and learn: Festschrift in honor of Lars Borin / Editors: Elena Volodina, Dana Dannélls, Aleksandrs Berdicevskis, Markus Forsberg, Shafqat Virk. - Göteborg : Institutionen för svenska, flerspråkighet och språkteknologi, Göteborgs universitet. - 1401-5919. - 9789187850837 ; , s. 29-37
-
Bokkapitel (refereegranskat)abstract
- FrameNet (FN) resources have existed for many languages for over a decade but their adoption in real world applications has been limited. To celebrate the 65 anniversary of Lars Borin, the initiator and leader of Swedish FrameNet, among others, we take a standpoint to motivate why language resources are crucial for moving NLP forward. We present our position on (a) the need for language resources to embrace other dimensions of text and language use, and (b) the need for them to relate to other representations through multimodality.
|
|
| 17. |
- de Graaf, Erik Wouter, 1991, et al.
(författare)
-
KILLE: Learning Objects and Spatial Relations with Kinect
- 2015
-
Ingår i: Proceedings of goDIAL - Semdial 2015: The 19th Workshop on the Semantics and Pragmatics of Dialogue. Gothenburg, 24–26 August 2015 / Christine Howes and Staffan Larsson (eds.). - 2308-2275.
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- We present a situated dialogue system designed to learn objects and spatial relations from relatively few examples, based on camera imagery and dialogue interaction with a human partner. We also report on the baseline evaluation of the system.
|
|
| 18. |
- Dobnik, Simon, 1977, et al.
(författare)
-
A formal semantic model for spatial descriptions
- 2015
-
Ingår i: Proceeds of the Workshop Formal Semantics Meets Cognitive Semantics, January 22-23 2015, Radboud University Nijmegen, The Netherlands.
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- In this presentation we discuss how our empirical investigations of learning geometric meanings of spatial descriptions with situated robots, learning functional meanings of prepositions from collections of image descriptions, and modelling of reference frame assignment in conversation can be formulated in the TTR framework. The overall goal is to provide an account of semantics of spatial prepositions for these modalities, and over longer term, use the framework as a knowledge representation system of a situated agent.
|
|
| 19. |
- Dobnik, Simon, 1977, et al.
(författare)
-
A Model for Attention-Driven Judgements in Type Theory with Records
- 2016
-
Ingår i: JerSem: The 20th Workshop on the Semantics and Pragmatics of Dialogue. J. Hunter, M. Simons, and M. Stone (eds.). New Brunswick, NJ USA, July 16–18 2016.. - 2308-2275.
-
Konferensbidrag (refereegranskat)abstract
- This paper makes three contributions to the discussion on the applicability of Type Theory with Records (TTR) to embodied dialogue agents. First, it highlights the problem of type assignment or judgements in practical implementations which is re- source intensive. Second, it presents a judgement control mechanism, which con- sists of grouping of types into clusters or states by their thematic relations and selec- tion of types following two mechanisms inspired by the Load Theory of selective attention and cognitive control (Lavie et al., 2004), that addresses this problem. Third, it presents a computational frame- work, based on Bayesian inference, that offers a basis for future practical exper- imentation on the feasibility of the pro- posed approach.
|
|
| 20. |
|
|
| 21. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Atention-driven Perceptual Classification and Reasoning
- 2016
-
Ingår i: The Sixth Swedish language technology conference (SLTC), Umeå University, 17-18 November, 2016, (ed. Björklund, Johanna and Stymne, Sara). - Umeå : Umeå University.
-
Konferensbidrag (refereegranskat)abstract
- One of the central challenges for embodied and situated agents with limited memory and processing resources is how to process and react to a large amount of information that they receive continuously through their sensors. Secondly, a model of infor- mation fusion is also required: how perceptual and linguistic/conceptual information interact and drive their reasoning. We present a computational model for perceptual classification in probabilistic Type Theory with Records (TTR), a formal meaning representation layer for such agents. The model is based on the cognitive notion of attention.
|
|
| 22. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Back to the Future: Logic and Machine Learning
- 2017
-
Ingår i: Conference on Logic and Machine Learning in Natural Language (LaML).
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- In this paper we argue that since the beginning of the natural language processing or computational linguistics there has been a strong connection between logic and machine learning. First of all, there is something logical about language or linguistic about logic. Secondly, we argue that rather than distinguishing between logic and machine learning, a more useful distinction is between top-down approaches and data-driven approaches. Examining some recent approaches in deep learning we argue that they incorporate both properties and this is the reason for their very successful adoption to solve several problems within language technology.
|
|
| 23. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Categorisation of conversational games in free dialogue over spatial scenes
- 2018
-
Ingår i: Proceedings of AixDial - Semdial 2018: The 22st Workshop on the Semantics and Pragmatics of Dialogue, Aix-en-Provence, 8-10 November 2018 / Laurent Prévot, Magalie Ochs and Benoît Favre (eds.). - Aix-en-Provence, France : Semdial. - 2308-2275.
-
Konferensbidrag (refereegranskat)abstract
- We describe an extension of a corpus of dialogues over perceptual scenes with the annotation of conversational games in which particular interactive strategies are adopted by conversational partners which result in regularities of dialogue features. We hope these will be useful for computational modelling of perceptual dialogue.
|
|
| 24. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Changing perspective: Local alignment of reference frames in dialogue
- 2015
-
Ingår i: Proceedings of SemDial 2015 (goDIAL): The 19th Workshop on the Semantics and Pragmatics of Dialogue. Christine Howes and Staffan Larsson (eds.). Gothenburg, 24–26 August 2015. - 2308-2275. ; , s. 24-32
-
Konferensbidrag (refereegranskat)
|
|
| 25. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Changing perspective: Local alignment of reference frames in dialogue
- 2015
-
Ingår i: The 6th annual CLT workshop. Tollered, 20-22 October, 2015.
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- In this paper we examine how people negotiate, interpret and repair the frame of reference (FoR) in free dialogues discussing spatial scenes. We describe a pilot study in which participants are given different perspectives of the same scene and asked to locate several objects that are only shown on one of their pictures. This task requires participants to coordinate on FoR in order to identify the missing objects. Preliminary results indicate that conversational participants align locally on FoR but do not converge on a global frame of reference. Misunderstandings lead to clarification sequences in which participants shift the FoR. These findings have implications for situated dialogue systems.
|
|
| 26. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Chat-o-matic: an online chat tool for collecting conversations of situated dialogue
- 2022
-
Ingår i: Proceedings of the 26th Workshop on the Semantics and Pragmatics of Dialogue, August, 22-24, 2022, Dublin. - Poster Abstracts. - 2308-2275.
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- We present a demo of an online chat tool that we have developed to collect conversational data from situated dialogue called Semant-o-matic chat or Chat-o-matic. This allows us to setup data collection experiments both with targeted participants and with online crowd-sourcing platforms.
|
|
| 27. |
|
|
| 28. |
|
|
| 29. |
- Dobnik, Simon, 1977
(författare)
-
Coordinating spatial perspective in discourse
- 2012
-
Ingår i: Proceedings of the Workshop on Vision and Language 2012 (VL'12): The 2nd Annual Meeting of the EPSRC Network on Vision and Language.
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- We present results of an on-line data collection experiment where we investigate the assignment and coordination of spatial perspective between a pair of dialogue participants situated in a constrained virtual environment.
|
|
| 30. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Exploration of functional semantics of prepositions from corpora of descriptions of visual scenes
- 2014
-
Ingår i: Proceedings of the Workshop on Vision and Language 2014 (VL'14), 25th International Conference on Computational Linguistics (COLING 2014), Dublin, 23rd August 2014. ; , s. 33-37
-
Konferensbidrag (refereegranskat)abstract
- We present a method of extracting functional semantic knowledge from corpora of descriptions of visual scenes. Such knowledge is required for interpretation and generation of spatial descriptions in tasks such as visual search. We identify semantic classes of target and landmark objects related by each preposition by abstracting over WordNet taxonomy. The inclusion of such knowledge in visual search should equip robots with a better, more human-like spatial cognition.
|
|
| 31. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Exploring the Functional and Geometric Bias of Spatial Relations Using Neural Language Models
- 2018
-
Ingår i: Proceedings of SpLU 2018 at NAACL-HLT 2018, June 6, 2018 New Orleans, Louisiana / Parisa Kordjamshidi, Archna Bhatia, James Pustejovsky, Marie-Francine Moens (eds.). - New Orleans, Louisiana, USA : Association of Computational Linguistics (ACL). - 9781948087216
-
Konferensbidrag (refereegranskat)abstract
- The challenge for computational models of spatial descriptions for situated dialogue systems is the integration of information from different modalities. The semantics of spatial descriptions are grounded in at least two sources of information: (i) a geometric representation of space and (ii) the functional interaction of related objects that. We train several neural language models on descriptions of scenes from a dataset of image captions and examine whether the functional or geometric bias of spatial descriptions reported in the literature is reflected in the estimated perplexity of these models. The results of these experiments have implications for the creation of models of spatial lexical semantics for human-robot dialogue systems. Furthermore, they also provide an insight into the kinds of the semantic knowledge captured by neural language models trained on spatial descriptions, which has implications for image captioning systems.
|
|
| 32. |
|
|
| 33. |
- Dobnik, Simon, 1977
(författare)
-
Hallucinate or ground: how general or specific are object descriptions generated by a vision-and-language transformer?
- 2022
-
Ingår i: 9th Swedish Language Technology Conference (SLTC), Stockholm, 23–25 november 2022. - Stockholm, Sweden : The division of Speech, Music & Hearing and Språkbanken Tal, KTH Royal Institute of Technology.
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- How do we refer to scene entities in visual scenes? We explore reference and re-reference in two vision and language tasks and link them to a model of attention. We discuss our find- ings in relation to modelling situated interac- tion in grounded language models and situated dialogue systems.
|
|
| 34. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Human evaluation of robot-generated spatial descriptions
- 2010
-
Ingår i: Proceedings of the Workshop on Computational Models of Spatial Language Interpretation (CoSLI) at Spatial Cognition 2010. - Mt. Hood/Portland, Oregon, USA : CEUR Workshop Proceedings. - 1613-0073.
-
Konferensbidrag (refereegranskat)abstract
- We describe a system where the semantics of spatial referential expressions have been automatically learned by finding mappings between symbolic natural language descriptions of the environment and non-symbolic representations from the sensory data of a mobile robot used for localisation and map building (SLAM). Although the success of learning can be measured by examining classifier performance on held-out data, this does not in itself guarantee that the descriptions generated will be natural and informative for a human observer. In this paper we describe the results of an evaluation of our embodied robotic system by human observers.
|
|
| 35. |
- Dobnik, Simon, 1977, et al.
(författare)
-
In Search of Meaning and Its Representations for Computational Linguistics
- 2022
-
Ingår i: Proceedings of the 2022 CLASP Conference on (Dis)embodiment, Gothenburg and online 15–16 September 2022 / Simon Dobnik, Julian Grove and Asad Sayeed (eds.). - : Association for Computational Linguistics. - 2002-9764. - 9781955917674
-
Konferensbidrag (refereegranskat)abstract
- In this paper we examine different meaning representations that are commonly used in different natural language applications today and discuss their limits, both in terms of the aspects of the natural language meaning they are modelling and in terms of the aspects of the application for which they are used.
|
|
| 36. |
- Dobnik, Simon, 1977
(författare)
-
Interfacing Language, Spatial Perception and Cognition in Type Theory with Records
- 2015
-
Ingår i: ESSLLI proceedings of the TYTLES workshop on Type Theory and Lexical Semantics (Cooper, Retoré, eds) ESSLLI August 3-14 2015, Barcelona. ; , s. 1-7
-
Konferensbidrag (refereegranskat)abstract
- We argue that computational modelling of perception, action, language, and cognition introduces several requirements on a formal semantic theory and its practical implementations. Using examples of semantic representations of spatial descriptions we show how Type Theory with Records (TTR) satisfies these requirements. The advantage of truth being based on agent-relative judgements in TTR is crucial in this but practically it comes with a computational cost. We argue that the number of type judgements an agent has to make can be minimised by incorporating a cognitive notion of judgement that is driven by perceptual attention.
|
|
| 37. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Interfacing language, spatial perception and cognition in Type Theory with Records
- 2017
-
Ingår i: Journal of Language Modelling. - 2299-856X .- 2299-8470. ; 5:2, s. 273-301
-
Tidskriftsartikel (refereegranskat)abstract
- We argue that computational modelling of perception, action, language, and cognition introduces several requirements on a formal semantic theory and its practical implementations in situated dialogue agents. Using examples of semantic representations of spatial descriptions we show how Type Theory with Records (TTR) satisfies these requirements and represents a promising knowledge representation system for situated agents.
|
|
| 38. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Investigating the role of priming and alignment of perspective in dialogue
- 2013
-
Ingår i: Proceedings of the 17th Workshop on the Semantics and Pragmatics of Dialogue, Amsterdam, 16–18 December 2013, Amsterdam. - 2308-2275. ; 17, s. 182-184
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- We examine the alignment of the primed frame of reference (FoR) for spatial descriptions over several utterances of a situated dialogue. We confirm the tendency of FoR alignment and that the intrinsic FoR is the most popular one independent of the priming.
|
|
| 39. |
- Dobnik, Simon, 1977, et al.
(författare)
-
KILLE: a Framework for Situated Agents for Learning Language Through Interaction
- 2017
-
Ingår i: Linköping Electronic Conference Proceedings. - Linköpings universitet : Linköping University Electronic Press. - 1650-3686 .- 1650-3740. - 9789176856017
-
Konferensbidrag (refereegranskat)abstract
- We present KILLE, a framework for situated agents for learning language through interaction with its environment (perception) and with a human tutor (dialogue). We provide proof-of-concept evaluations of the usability of the system in two domains: learning of object categories and learning of spatial relations.
|
|
| 40. |
- Dobnik, Simon, 1977, et al.
(författare)
-
KILLE: learning grounded language through interaction
- 2017
-
Ingår i: CEUR Workshop Proceedings. - : CEUR-WS.org, urn:nbn:de:0074-1863-7. - 1613-0073.
-
Konferensbidrag (refereegranskat)abstract
- Testing and computational implementation of formal models of situated linguistic interaction imposes demands on computational infrastructure. We present our system called KILLE and provide a proof-of- concept evaluation of interactive situated learning of object categories and spatial relations.
|
|
| 41. |
- Dobnik, Simon, 1977
(författare)
-
Language, action, and perception
- 2018
-
Ingår i: Workshop on Annotation, Recognition and Evaluation of Actions (AREA) at LREC, 7 May 2018, Miyazaki, Japan / Pustejovsky, James and van der Sluis,Lelka (eds.). - Miyazaki, Japan : European Language Resources Association (ELRA). - 9791095546061
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- Situated agents interact both with their physical environment they are located in and with their conversational partners. As both the world and the language used in situated conversations are continuously changing, an agent must be able to adapt its grounded semantic representations by learning from new information. A pre-requisite for a dynamic, interactive approach to learning of grounded semantic representations is that an agent is equipped with a set of actions that define its strategies for identifying and connecting linguistic and perceptual information to its knowledge. In this talk we present our work on grounding spatial descriptions that argues that perceptual grounding is dynamic and adaptable to contexts. We describe a system called Kille which we use for interactive learning of objects and spatial relations from a human tutor. Finally, we describe our work on identifying interactive strategies of frame of reference assignment in spatial descriptions in a corpus of human-human dialogues and argue that there is no general preference for frame of reference assignment but this is linked to interaction strategies between agents that are adopted within a particular dialogue game.
|
|
| 42. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Language, Action and Perception (APL-ESSLLI): Lecture Notes of a Course in Language and Computation
- 2019
-
Ingår i: ESSLLI 2019, 31 European Summer School on Logic, Language and Information, 5 -6 August 2019, Riga, Latvia. - Riga, Latvia : University of Latvia.
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- The course gives a survey of theory and practical computational implementations of how natural language interacts with the physical world through action and perception. We will look at topics such as semantic theories and computational approaches to modelling natural language, action and perception (grounding), situated dialogue systems, integrated robotic systems, grounding of language in action and perception, generation and interpretation of scene descriptions from images and videos, spatial cognition, and others.
|
|
| 43. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Local alignment of frame of reference assignment in English and Swedish dialogue
- 2020
-
Ingår i: Spatial Cognition XII, 12th International Conference, Spatial Cognition 2020, Riga, Latvia, August 26–28, 2020, Proceedings / Jurǵis Šķilters, Nora S. Newcombe, David Uttal (eds.). - Cham : Springer Nature Switzerland AG. - 0302-9743 .- 1611-3349. - 9783030579821
-
Konferensbidrag (refereegranskat)abstract
- In this paper we examine how people assign, interpret, negotiate and repair the frame of reference (FoR) in online text-based dialogues discussing spatial scenes in English and Swedish. We describe our corpus and data collection which involves a coordination experiment in which dyadic dialogue participants have to identify differences in their picture of a visual scene. As their perspectives of the scene are different, they must coordinate their FoRs in order to complete the task. Results show that participants do not align on a global FoR, but tend to align locally, for sub-portions (or particular conversational games) in the dialogue. This has implications for how dialogue systems should approach problems of FoR assignment - and what strategies for clarification they should implement.
|
|
| 44. |
- Dobnik, Simon, 1977
(författare)
-
Mastering language and computation: how well is teaching of language technology adapted to student learning?
- 2019
-
Ingår i: Internal report. - Gothenburg, Sweden : University of Gothenburg, Pedagogical Development and Interactive Learning (PIL).
-
Annan publikation (övrigt vetenskapligt/konstnärligt)abstract
- In this paper we investigate self-reported knowledge and skills of students in the interdisciplinary masters programme in language technology where the majority of students have only studied one discipline at the previous level. We focus on the answers of students in Year 1 of the programme and those in Year 2 in order to examine how well are the current teaching strategies adapted to decrease this gap. Comparing the results with the individual reflections of the author of this study, who is a teacher in this programme, we make recommendations about improving the teaching strategies in the programme.
|
|
| 45. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Modelling language, action and perception in Type Theory with Records : Language, action and perception in TTR
- 2012
-
Ingår i: Proceedings of the 7th International Workshop on Constraint Solving and Language Processing (CSLP'12). ; , s. 51-62
-
Konferensbidrag (refereegranskat)abstract
- We present a formal model for natural language semantics using Type Theory with Records (TTR) and argue that it is better suited for representing the meaning of spatial descriptions than traditional for- mal semantic models. Spatial descriptions include perceptual, conceptual and discourse knowledge which we represent all in a single framework. Being a computational framework TTR is suited for modelling language and cognition of conversational agents in robotics and virtual environ- ments where interoperability between language, action and perception is required. The perceptual systems gain access to abstract conceptual meaning representations of language while the latter can be justified in action and perception.
|
|
| 46. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Modelling Language, Action, and Perception in Type Theory with Records
- 2013
-
Ingår i: Constraint Solving and Language Processing: 7th International Workshop, CSLP 2012, Orléans, France, September 13-14, 2012, Revised Selected Papers / Editors: Denys Duchier, Yannick Parmentier. - Berlin Heidelberg : Springer-Verlag. - 9783642415777 ; , s. 70-91
-
Bokkapitel (refereegranskat)abstract
- Formal models of natural language semantics using TTR (Type Theory with Records) attempt to relate natural language to perception, modelled as classification of objects and events by types which are available as resources to an agent. We argue that this is better suited for representing the meaning of spatial descriptions in the context of agent modelling than traditional formal semantic models which do not relate spatial concepts to perceptual apparatus. Spatial descriptions include perceptual, conceptual and discourse knowledge which we represent all in a single framework. Being a general framework for modelling both linguistic and non-linguistic cognition, TTR is more suitable for the modelling of situated conversational agents in robotics and virtual environments where interoperability between language, action and perception is required. The perceptual systems gain access to abstract conceptual meaning representations of language while the latter can be justified in action and perception.
|
|
| 47. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Modular Mechanistic Networks for Computational Modelling of Spatial Descriptions
- 2018
-
Ingår i: Abstracts of the Seventh Swedish Language Technology Conference (SLTC-2018), 7-9 November 2018, Stockholm. - Stockholm, Sweden.
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- We argue current deep learning approaches to modelling of spatial language in generating image captions have shortcomings and that the multiplicity of factors that influence spatial language invites a modular approach where the solution can be built in a piece-wise manner and then integrated. We call this approach where deep learning is assisted with domain knowledge expressed as modules that are trained on data a top-down or mechanistic approach to otherwise a bottom-up phenomenological approach.
|
|
| 48. |
- Dobnik, Simon, 1977, et al.
(författare)
-
Modular mechanistic networks: On bridging mechanistic and phenomenological models with deep neural networks in natural language processing
- 2017
-
Ingår i: CLASP Papers in Computational Linguistics: Proceedings of the Conference on Logic and Machine Learning in Natural Language (LaML 2017), Gothenburg, 12–13 June 2017 (ISSN application pending). - Gothenburg, Sweden : CLASP: Centre for Language and Studies in Probability, FLOV, University of Gothenburg. - 2002-9764.
-
Konferensbidrag (refereegranskat)abstract
- Natural language processing (NLP) can be done using either top-down (theory driven) and bottom-up (data driven) approaches, which we call mechanistic and phenomenological respectively. The approaches are frequently considered to stand in opposition to each other. Examining some recent approaches in deep learning we argue that deep neural networks incorporate both perspectives and, furthermore, that leveraging this aspect of deep learning may help in solving complex problems within language technology, such as modelling language and perception in the domain of spatial cognition.
|
|
| 49. |
- Dobnik, Simon, 1977
(författare)
-
Neat and Scruffy: On Computational Generation and Interpretation of Spatial Descriptions
- 2022
-
Ingår i: Proceedings of the 1st International Workshop on Spatio-Temporal Reasoning and Learning (STRL 2022) co-located with the 31st International Joint Conference on Artificial Intelligence and the 25th European Conference on Artificial Intelligence (IJCAI 2022, ECAI 2022), Vienna, Austria, July 24, 2022.. - Vienna, Austria : CEUR Workshop Proceedings (CEUR-WS.org). - 1613-0073.
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)
|
|
| 50. |
- Dobnik, Simon, 1977
(författare)
-
O prijaznih računalnikih, ki nas razumejo in se od nas učijo : About friendly computers that understand and learn from us
- 2019
-
Ingår i: Invited talk: KAZU, Koroška akademija znanosti in umetnosti (Carinthian academy of science and art).
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- Računalniška lingvistika, jezikovna tehnologija ali procesiranje naravnega jezika je tisti del umetne inteligence, ki se ukvarja z modeliranjem človeškega ali naravnega jezika, tako besed kot govora, z računalniškimi programi s ciljem, da bi ga ti proizvedli ali razumeli na enka način kot pa ljudje. To je pomembno iz dveh razlogov. Velik del našega znanja je zapisan prav v naravnem jeziku, na primer v besedilih, ki jih najdemo na spletu, in iz katerih želimo izluščiti pomen, jih povzeti ali prevesti v drug jezik. Jezik je tudi najbolj naraven način komunikacije med nami, vendar večina ni vešča računalniškega programiranja. Namesto, da se vsak od nas nauči računalniškega programskega jezika, je veliko bolj naravno, da se s pametnim asistentom ali robotom sporazumevamo kar v slovenščini in ga tako tudi “programiramo”. Preden se lotimo dela, je potrebno, da vemo nekaj o samem jeziku. Kako deluje? Kako razumemo pomen besed in stavkov? Kakšna so pravila sporazumevanja med nami? Ljudje in računalniki (tudi najsodobnejši roboti) smo si zelo različni, zato moramo najti načine (algoritme ali sisteme), ki poustvarijo jezikovno vedenje navzven, ki je podobno našemu. V ozadju si pomagamo z matematiko in logiko, programskimi jeziki, teorijo verjetnosti, strojnim učenjem, rudarjenjem podatkov in še čem. Računalniki programi se učijo iz vzorcev informacij, ki jim jih posredujemo, vendar pa pogosto to niso enake informacije – zvoki in podobe, ki jih vidi otrok, ko se uči jezika. Prav tako se računalniki ne učijo jezika v sporazumevanju z drugimi ljudmi več let. Ljudje lahko tudi hitro prilagodimo pomen in izbiro besed glede na različne situacije, vključno s tistimi, ko se sporazumevamo z računalniki. Torej morajo biti računalniški programi enako prilagodljivi do nas in se učiti od nas. Delo z umetno inteligentninimi sistemi pa nam ne prinese le novih dognanj, kako izboljšati jezikovne sposobnosti računalnikov, temveč nam da tudi odgovore o tem, kako posebna je človeška inteligenca.
|
|
