Publications

This chapter demonstrates how the Royal Society Corpus, a richly annotated corpus of around 48,000 English scientific journal articles covering more than 330 years, can be used for lexico-grammatical and pragmatic studies that contribute to a broader understanding of the development of medical research articles. The Late Modern English period together with several decades before and after this time frame was a productive period in the medical output of the Royal Society. This chapter addresses typical linguistic features of scientific journal articles from medical and related sciences from this period demonstrating their special status in the context of other traditional and emerging disciplines in the corpus data. Additionally, language usage and text-type conventions of historical medical research articles will be compared to the features of corpus texts on medical topics from Present-day English.

• https://benjamins.com/catalog/pbns.330

@inbook{MedicalDiscourse22,
title = {Medical discourse in Late Modern English: Insights from the Royal Society Corpus.},
author = {Katrin Menzel},
editor = {Turo Hiltunen and Irma Taavitsainen},
url = {https://benjamins.com/catalog/pbns.330},
year = {2022},
date = {2022},
booktitle = {Corpus pragmatic studies on the history of medical discourse (Pragmatics & Beyond New Series; Vol. 330)},
pages = {79-104},
publisher = {John Benjamins},
address = {Amsterdam},
abstract = {This chapter demonstrates how the Royal Society Corpus, a richly annotated corpus of around 48,000 English scientific journal articles covering more than 330 years, can be used for lexico-grammatical and pragmatic studies that contribute to a broader understanding of the development of medical research articles. The Late Modern English period together with several decades before and after this time frame was a productive period in the medical output of the Royal Society. This chapter addresses typical linguistic features of scientific journal articles from medical and related sciences from this period demonstrating their special status in the context of other traditional and emerging disciplines in the corpus data. Additionally, language usage and text-type conventions of historical medical research articles will be compared to the features of corpus texts on medical topics from Present-day English.},
pubstate = {published},
type = {inbook}
}

Copy BibTeX to Clipboard

Plan Verification is the task of deciding whether a sequence of actions is a solution for a given planning problem. In HTN planning, the task is computationally expensive and may be up to NP-hard. However, there are situations where it needs to be solved, e.g. when a solution is post-processed, in systems using approximation, or just to validate whether a planning system works correctly (e.g. for debugging or in a competition). There are verification systems based on translations to propositional logic and on techniques from parsing. Here we present a third approach and translate HTN plan verification problems into HTN planning problems. These can be solved using any HTN planning system. We collected a new bench-mark set based on models and results of the 2020 International Planning Competition. Our evaluation shows that our compilation outperforms the approaches from the literature.

• 19795-Article Text-23808-1-2-20220613     (0.39MB)
• https://ojs.aaai.org/index.php/ICAPS/article/view/19795/19554
• https://doi.org/10.1609/icaps.v32i1.19795

@inproceedings{Höller_Wichlacz_Bercher_Behnke_2022,
title = {Compiling HTN Plan Verification Problems into HTN Planning Problems},
author = {Daniel H{\"o}ller and Julia Wichlacz and Pascal Bercher and Gregor Behnke},
url = {https://ojs.aaai.org/index.php/ICAPS/article/view/19795/19554},
doi = {https://doi.org/10.1609/icaps.v32i1.19795},
year = {2022},
date = {2022},
booktitle = {Proceedings of the Thirty-Second International Conference on Automated Planning and Scheduling (ICAPS2022)},
pages = {145-150},
abstract = {Plan Verification is the task of deciding whether a sequence of actions is a solution for a given planning problem. In HTN planning, the task is computationally expensive and may be up to NP-hard. However, there are situations where it needs to be solved, e.g. when a solution is post-processed, in systems using approximation, or just to validate whether a planning system works correctly (e.g. for debugging or in a competition). There are verification systems based on translations to propositional logic and on techniques from parsing. Here we present a third approach and translate HTN plan verification problems into HTN planning problems. These can be solved using any HTN planning system. We collected a new bench-mark set based on models and results of the 2020 International Planning Competition. Our evaluation shows that our compilation outperforms the approaches from the literature.},
pubstate = {published},
type = {inproceedings}
}

Copy BibTeX to Clipboard

While state-of-the-art planning systems need a grounded (propositional) task representation, the input model is provided “lifted”, specifying predicates and action schemas with variables over a finite object universe. The size of the grounded model is exponential in predicate/action-schema arity, limiting applicability to cases where it is small enough. Recent work has taken up this challenge, devising an effective lifted forward search planner as basis for lifted heuristic search, as well as a variety of lifted heuristic functions based on the delete relaxation. Here we add a novel family of lifted heuristic functions, based on landmarks. We design two methods for landmark extraction in the lifted setting. The resulting heuristics exhibit performance advantages over previous heuristics in several benchmark domains. Especially the combination with lifted delete relaxation heuristics to a LAMA-style planner yields good results, beating the previous state of the art in lifted planning.

• Landmark-Heuristics-for-Lifted-Classical-Planning     (0.2MB)
• https://doi.org/10.24963/ijcai.2022/647
• https://doi.org/10.24963/ijcai.2022/647

@inproceedings{ijcai2022p647,
title = {Landmark Heuristics for Lifted Classical Planning},
author = {Julia Wichlacz and Daniel H{\"o}ller and J{\"o}rg Hoffmann},
editor = {Lud De Raedt},
url = {https://doi.org/10.24963/ijcai.2022/647},
doi = {https://doi.org/10.24963/ijcai.2022/647},
year = {2022},
date = {2022},
booktitle = {Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence (IJCAI-22), Vienna, 23-29 July 2022},
pages = {4665-4671},
publisher = {International Joint Conferences on Artificial Intelligence Organization},
abstract = {While state-of-the-art planning systems need a grounded (propositional) task representation, the input model is provided “lifted”, specifying predicates and action schemas with variables over a finite object universe. The size of the grounded model is exponential in predicate/action-schema arity, limiting applicability to cases where it is small enough. Recent work has taken up this challenge, devising an effective lifted forward search planner as basis for lifted heuristic search, as well as a variety of lifted heuristic functions based on the delete relaxation. Here we add a novel family of lifted heuristic functions, based on landmarks. We design two methods for landmark extraction in the lifted setting. The resulting heuristics exhibit performance advantages over previous heuristics in several benchmark domains. Especially the combination with lifted delete relaxation heuristics to a LAMA-style planner yields good results, beating the previous state of the art in lifted planning.},
pubstate = {published},
type = {inproceedings}
}

Copy BibTeX to Clipboard

Script knowledge (Schank and Abelson, 1977) is useful for a variety of NLP tasks. However, existing resources only cover a small number of activities, limiting its practical usefulness. In this work, we propose a zero-shot learning approach to script parsing, the task of tagging texts with scenario-specific event and participant types, which enables us to acquire script knowledge without domain-specific annotations. We (1) learn representations of potential event and participant mentions by promoting class consistency according to the annotated data; (2) perform clustering on the event /participant candidates from unannotated texts that belongs to an unseen scenario. The model achieves 68.1/74.4 average F1 for event / participant parsing, respectively, outperforming a previous CRF model that, in contrast, has access to scenario-specific supervision. We also evaluate the model by testing on a different corpus, where it achieved 55.5/54.0 average F1 for event / participant parsing.

• 2022.coling-1.356     (0.63MB)
• https://aclanthology.org/2022.coling-1.356

@inproceedings{zhai-etal-2022-zero,
title = {Zero-shot Script Parsing},
author = {Fangzhou Zhai and Vera Demberg and Alexander Koller},
url = {https://aclanthology.org/2022.coling-1.356},
year = {2022},
date = {2022},
booktitle = {Proceedings of the 29th International Conference on Computational Linguistics},
pages = {4049-4060},
publisher = {International Committee on Computational Linguistics},
address = {Gyeongju, Republic of Korea},
abstract = {Script knowledge (Schank and Abelson, 1977) is useful for a variety of NLP tasks. However, existing resources only cover a small number of activities, limiting its practical usefulness. In this work, we propose a zero-shot learning approach to script parsing, the task of tagging texts with scenario-specific event and participant types, which enables us to acquire script knowledge without domain-specific annotations. We (1) learn representations of potential event and participant mentions by promoting class consistency according to the annotated data; (2) perform clustering on the event /participant candidates from unannotated texts that belongs to an unseen scenario. The model achieves 68.1/74.4 average F1 for event / participant parsing, respectively, outperforming a previous CRF model that, in contrast, has access to scenario-specific supervision. We also evaluate the model by testing on a different corpus, where it achieved 55.5/54.0 average F1 for event / participant parsing.},
pubstate = {published},
type = {inproceedings}
}

Copy BibTeX to Clipboard

@inbook{Zygis2021intonation,
title = {Intonation: Other pragmatic functions and phonetic / phonological effects},
author = {Marzena Żygis and Štefan Beňuš and Bistra Andreeva},
editor = {Neil Bermel and Jan Fellerer},
year = {2022},
date = {2022},
booktitle = {The Oxford Guide to the Slavonic Languages},
publisher = {Oxford University Press},
pubstate = {published},
type = {inbook}
}

Copy BibTeX to Clipboard

@inbook{Andreeva2022intonation,
title = {Intonation and information structure},
author = {Bistra Andreeva and Snezhina Dimitrova},
editor = {Neil Bermel and Jan Fellerer},
year = {2022},
date = {2022},
booktitle = {The Oxford Guide to the Slavonic Languages},
publisher = {Oxford University Press},
pubstate = {forthcoming},
type = {inbook}
}

Copy BibTeX to Clipboard

One reason why intertwined clauses are harder to perceive than their counterparts might be information structure (IS). The current idea is that speakers choose extraposition of new information to help the audience perceive their text and to distribute information evenly across a discourse (Levy and Jaeger 2007). The purpose of this paper is to verify the relation between IS and extraposition or
adjacency of relative clauses (RC).

• http://www.dgfs2019.uni-bremen.de/abstracts/ag7/Voigtmann.pdf

@inproceedings{voigtmanninprint,
title = {Informational aspects of the extraposition of relative clauses},
author = {Sophia Voigtmann},
editor = {Marco Coniglio and Chiara de Bastiani and Nicholas Catasso},
url = {http://www.dgfs2019.uni-bremen.de/abstracts/ag7/Voigtmann.pdf},
year = {2022},
date = {2022},
booktitle = {Language Change at the Interfaces. Intrasentential and intersentential phenomena},
abstract = {One reason why intertwined clauses are harder to perceive than their counterparts might be information structure (IS). The current idea is that speakers choose extraposition of new information to help the audience perceive their text and to distribute information evenly across a discourse (Levy and Jaeger 2007). The purpose of this paper is to verify the relation between IS and extraposition or adjacency of relative clauses (RC).},
pubstate = {published},
type = {inproceedings}
}

Copy BibTeX to Clipboard

In the theoretical literature, the unacceptability of (some) structural mismatches between the antecedent and the target of ellipsis have been taken to indicate that ellipsis is subject to syntactic identity conditions. Such constraints have been defended for verb phrase ellipsis (VPE) (Arregui et al., 2006; Merchant, 2013) and sluicing (Chung, 2006, 2013). The assumption of syntactic identity conditions increases the complexity of the grammar, because conditions which are specific to particular ellipses must be added to a system of more general rules. If the data that apparently support syntactic identity conditions could be explained by independently motivated principles, this would consequently reduce the complexity of the syntactic system. In this article we investigate syntactic identity conditions proposed by Chung (2006, 2013) for sluicing, i.e. the ellipsis of the TP in a wh-question, which is survived only by the wh-phrase (1a) (Ross, 1969). Our study shows that apparent grammaticality contrasts can be accounted for by a probabilistic processing account, which is supported by an acceptability rating, a production and a self-paced reading experiment. In contrast, Chung’s constraints lead to predictions which are not supported by our data.

• LE20Procs_#29_Lemke_541     (2.24MB)
• https://publikationen.uni-tuebingen.de/xmlui/handle/10900/119301

@inproceedings{lemke.etalidentity,
title = {Can identity conditions on ellipsis be explained by processing principles?},
author = {Tyll Robin Lemke and Lisa Sch{\"a}fer and Ingo Reich},
editor = {Robin H{\"o}rnig and Sophie von Wietersheim and Andreas Konietzko and Sam Featherston},
url = {https://publikationen.uni-tuebingen.de/xmlui/handle/10900/119301},
year = {2022},
date = {2022},
booktitle = {Proceedings of Linguistic Evidence 2020: Linguistic Theory Enriched by Experimental Data},
pages = {541-561},
publisher = {University of T{\"u}bingen},
address = {T{\"u}bingen, Germany},
abstract = {In the theoretical literature, the unacceptability of (some) structural mismatches between the antecedent and the target of ellipsis have been taken to indicate that ellipsis is subject to syntactic identity conditions. Such constraints have been defended for verb phrase ellipsis (VPE) (Arregui et al., 2006; Merchant, 2013) and sluicing (Chung, 2006, 2013). The assumption of syntactic identity conditions increases the complexity of the grammar, because conditions which are specific to particular ellipses must be added to a system of more general rules. If the data that apparently support syntactic identity conditions could be explained by independently motivated principles, this would consequently reduce the complexity of the syntactic system. In this article we investigate syntactic identity conditions proposed by Chung (2006, 2013) for sluicing, i.e. the ellipsis of the TP in a wh-question, which is survived only by the wh-phrase (1a) (Ross, 1969). Our study shows that apparent grammaticality contrasts can be accounted for by a probabilistic processing account, which is supported by an acceptability rating, a production and a self-paced reading experiment. In contrast, Chung’s constraints lead to predictions which are not supported by our data.},
pubstate = {published},
type = {inproceedings}
}

Copy BibTeX to Clipboard

@incollection{lemke.etalquestions,
title = {Questions under discussion, salience and the acceptability of fragments},
author = {Tyll Robin Lemke and Ingo Reich and Lisa Sch{\"a}fer},
editor = {Andreas Konietzko and Susanne Winkler},
year = {2022},
date = {2022},
booktitle = {Information Structure and Discourse in Generative Grammar: Mechanisms and Processes},
publisher = {De Gruyter Mouton},
address = {Berlin; Boston},
pubstate = {forthcoming},
type = {incollection}
}

Copy BibTeX to Clipboard

Decades of studies trying to define the extent to which artificial neural networks can exhibit systematicity suggest that systematicity can be achieved by connectionist models but not by default. Here we present a novel connectionist model of sentence production that employs rich situation model representations originally proposed for modeling systematicity in comprehension. The high performance of our model demonstrates that such representations are also well suited to model language production. Furthermore, the model can produce multiple novel sentences for previously unseen situations, including in a different voice (actives vs. passive) and with words in new syntactic roles, thus demonstrating semantic and syntactic generalization and arguably systematicity. Our results provide yet further evidence that such connectionist approaches can achieve systematicity, in production as well as comprehension. We propose our positive results to be a consequence of the regularities of the microworld from which the semantic representations are derived, which provides a sufficient structure from which the neural network can interpret novel inputs.

• https://doi.org/10.3390/info12080329
• https://doi.org/10.3390/info12080329

@article{Calvillo2021semantic,
title = {Semantic Systematicity in Connectionist Language Production},
author = {Jesús Calvillo and Harm Brouwer and Matthew W. Crocker},
url = {https://doi.org/10.3390/info12080329},
doi = {https://doi.org/10.3390/info12080329},
year = {2021},
date = {2021},
journal = {Information (Special issue on Neural Natural Language Generation)},
pages = {329},
volume = {12},
number = {8},
abstract = {Decades of studies trying to define the extent to which artificial neural networks can exhibit systematicity suggest that systematicity can be achieved by connectionist models but not by default. Here we present a novel connectionist model of sentence production that employs rich situation model representations originally proposed for modeling systematicity in comprehension. The high performance of our model demonstrates that such representations are also well suited to model language production. Furthermore, the model can produce multiple novel sentences for previously unseen situations, including in a different voice (actives vs. passive) and with words in new syntactic roles, thus demonstrating semantic and syntactic generalization and arguably systematicity. Our results provide yet further evidence that such connectionist approaches can achieve systematicity, in production as well as comprehension. We propose our positive results to be a consequence of the regularities of the microworld from which the semantic representations are derived, which provides a sufficient structure from which the neural network can interpret novel inputs.},
pubstate = {published},
type = {article}
}

Copy BibTeX to Clipboard

• https://doi.org/10.1017/S0305000920000227
• https://doi.org/10.1017/S0305000920000227

@article{YUENetal2020cues,
title = {Five-year-olds produce prosodic cues to distinguish compounds from lists in Australian English},
author = {Ivan Yuen and Nan Xu Rattanasone and Elaine Schmidt and Gretel Macdonald and Rebecca Holt and Katherine Demuth},
url = {https://doi.org/10.1017/S0305000920000227},
doi = {https://doi.org/10.1017/S0305000920000227},
year = {2021},
date = {2021},
journal = {Journal of Child Language},
pages = {110-128},
publisher = {Cambridge University Press},
volume = {48},
number = {1},
abstract = {

Copy BibTeX to Clipboard

We present a Wizard-of-Oz experiment examining phonetic accommodation of human interlocutors in the context of human-computer interaction. Forty-two native speakers of German engaged in dynamic spoken interaction with a simulated virtual tutor for learning the German language called Mirabella. Mirabella was controlled by the experimenter and used either natural or hidden Markov model-based synthetic speech to communicate with the participants. In the course of four tasks, the participants’ accommodating behavior with respect to wh-question realization and allophonic variation in German was tested. The participants converged to Mirabella with respect to modified wh-question intonation, i.e., rising F₀ contour and nuclear pitch accent on the interrogative pronoun, and the allophonic contrast [ɪç] vs. [ɪk] occurring in the word ending 〈-ig〉. They did not accommodate to the allophonic contrast [ɛː] vs. [eː] as a realization of the long vowel 〈-ä-〉. The results did not differ between the experimental groups that communicated with either the natural or the synthetic speech version of Mirabella. Testing the influence of the “Big Five” personality traits on the accommodating behavior revealed a tendency for neuroticism to influence the convergence of question intonation. On the level of individual speakers, we found considerable variation with respect to the degree and direction of accommodation. We conclude that phonetic accommodation on the level of local prosody and segmental pronunciation occurs in users of spoken dialog systems, which could be exploited in the context of computer-assisted language learning.

• https://doi.org/10.1016/j.wocn.2021.101029
• https://doi.org/10.1016/j.wocn.2021.101029

@article{Gessinger/etal:2021a,
title = {Phonetic accommodation in interaction with a virtual language learning tutor: A Wizard-of-Oz study},
author = {Iona Gessinger and Bernd M{\"o}bius and S{\'e}bastien Le Maguer and Eran Raveh and Ingmar Steiner},
url = {https://doi.org/10.1016/j.wocn.2021.101029},
doi = {https://doi.org/10.1016/j.wocn.2021.101029},
year = {2021},
date = {2021},
journal = {Journal of Phonetics},
pages = {101029},
volume = {86},
abstract = {We present a Wizard-of-Oz experiment examining phonetic accommodation of human interlocutors in the context of human-computer interaction. Forty-two native speakers of German engaged in dynamic spoken interaction with a simulated virtual tutor for learning the German language called Mirabella. Mirabella was controlled by the experimenter and used either natural or hidden Markov model-based synthetic speech to communicate with the participants. In the course of four tasks, the participants’ accommodating behavior with respect to wh-question realization and allophonic variation in German was tested. The participants converged to Mirabella with respect to modified wh-question intonation, i.e., rising F₀ contour and nuclear pitch accent on the interrogative pronoun, and the allophonic contrast [ɪç] vs. [ɪk] occurring in the word ending 〈-ig〉. They did not accommodate to the allophonic contrast [ɛː] vs. [eː] as a realization of the long vowel 〈-{\"a}-〉. The results did not differ between the experimental groups that communicated with either the natural or the synthetic speech version of Mirabella. Testing the influence of the “Big Five” personality traits on the accommodating behavior revealed a tendency for neuroticism to influence the convergence of question intonation. On the level of individual speakers, we found considerable variation with respect to the degree and direction of accommodation. We conclude that phonetic accommodation on the level of local prosody and segmental pronunciation occurs in users of spoken dialog systems, which could be exploited in the context of computer-assisted language learning.},
pubstate = {published},
type = {article}
}

Copy BibTeX to Clipboard

• https://aclanthology.org/2021.motra-1.0/

@proceeding{motra-2021-modelling,
title = {Proceedings for the First Workshop on Modelling Translation: Translatology in the Digital Age},
author = {},
editor = {Yuri Bizzoni and Elke Teich and Cristina Espa{\~n}a-Bonet and Josef van Genabith},
url = {https://aclanthology.org/2021.motra-1.0/},
year = {2021},
date = {2021},
publisher = {Association for Computational Linguistics},
address = {online},
pubstate = {published},
type = {proceeding}
}

Copy BibTeX to Clipboard

While some authors have suggested that translationese fingerprints are universal, others have shown that there is a fair amount of variation among translations due to source language shining through, translation type or translation mode. In our work, we attempt to gain empirical insights into variation in translation, focusing here on translation mode (translation vs. interpreting). Our goal is to discover features of translationese and interpretese that distinguish translated and interpreted output from comparable original text/speech as well as from each other at different linguistic levels. We use relative entropy (Kullback-Leibler Divergence) and visualization with word clouds. Our analysis shows differences in typical words between originals vs. non-originals as well as between translation modes both at lexical and grammatical levels.

• https://doi.org/10.1075/btl.158.12kar
• https://doi.org/10.1075/btl.158.12kar

@incollection{KarakantaEtAl2021,
title = {Exploring variation in translation with probabilistic language models},
author = {Alina Karakanta and Heike Przybyl and Elke Teich},
editor = {Julia Lavid-López and Carmen Ma{\'i}z-Ar{\'e}valo and Juan Rafael Zamorano-Mansilla},
url = {https://doi.org/10.1075/btl.158.12kar},
doi = {https://doi.org/10.1075/btl.158.12kar},
year = {2021},
date = {2021},
booktitle = {Corpora in Translation and Contrastive Research in the Digital Age: Recent advances and explorations},
pages = {308-323},
publisher = {Benjamins},
address = {Amsterdam},
abstract = {While some authors have suggested that translationese fingerprints are universal, others have shown that there is a fair amount of variation among translations due to source language shining through, translation type or translation mode. In our work, we attempt to gain empirical insights into variation in translation, focusing here on translation mode (translation vs. interpreting). Our goal is to discover features of translationese and interpretese that distinguish translated and interpreted output from comparable original text/speech as well as from each other at different linguistic levels. We use relative entropy (Kullback-Leibler Divergence) and visualization with word clouds. Our analysis shows differences in typical words between originals vs. non-originals as well as between translation modes both at lexical and grammatical levels.},
pubstate = {published},
type = {incollection}
}

Copy BibTeX to Clipboard

In the present paper, we explore lexical contexts of discourse markers in translation and interpreting on the basis of word embeddings. Our special interest is on contextual variation of the same discourse markers in (written) translation vs. (simultaneous) interpreting. To explore this variation at the lexical level, we use a data-driven approach: we compare bilingual neural word embeddings trained on source-to- translation and source-tointerpreting aligned corpora. Our results show more variation of semantically related items in translation spaces vs. interpreting ones and a more consistent use of fewer connectives in interpreting. We also observe different trends with regard to the discourse relation types.

• 2021.codi-main.13     (0.19MB)
• https://aclanthology.org/2021.codi-main.13/

@inproceedings{LapshinovaEtAl2021codi,
title = {Tracing variation in discourse connectives in translation and interpreting through neural semantic spaces},
author = {Ekaterina Lapshinova-Koltunski and Heike Przybyl and Yuri Bizzoni},
url = {https://aclanthology.org/2021.codi-main.13/},
year = {2021},
date = {2021},
booktitle = {Proceedings of the 2nd Workshop on Computational Approaches to Discourse CODI},
pages = {134-142},
address = {Punta Cana, Dominican Republic and Online},
abstract = {In the present paper, we explore lexical contexts of discourse markers in translation and interpreting on the basis of word embeddings. Our special interest is on contextual variation of the same discourse markers in (written) translation vs. (simultaneous) interpreting. To explore this variation at the lexical level, we use a data-driven approach: we compare bilingual neural word embeddings trained on source-to- translation and source-tointerpreting aligned corpora. Our results show more variation of semantically related items in translation spaces vs. interpreting ones and a more consistent use of fewer connectives in interpreting. We also observe different trends with regard to the discourse relation types.},
pubstate = {published},
type = {inproceedings}
}

Copy BibTeX to Clipboard

Translationese data is a scarce and valuable resource. Traditionally, the proceedings of the European Parliament have been used for studying translationese phenomena since their metadata allows to distinguish between original and translated texts. However, translations are not always direct and we hypothesise that a pivot (also called ”relay”) language might alter the conclusions on translationese effects. In this work, we (i) isolate translations that have been done without an intermediate language in the Europarl proceedings from those that might have used a pivot language, and (ii) build comparable and parallel corpora with data aligned across multiple languages that therefore can be used for both machine translation and translation studies.

• https://aclanthology.org/2021.motra-1.1/

@inproceedings{AmposahEtal:MOTRA:2021,
title = {Do not Rely on Relay Translations: Multilingual Parallel Direct Europarl},
author = {Kwabena Amponsah-Kaakyire and Daria Pylypenko and Cristina Espa{\~n}a-Bonet and Josef van Genabith},
url = {https://aclanthology.org/2021.motra-1.1/},
year = {2021},
date = {2021},
booktitle = {Proceedings of the Workshop on Modelling Translation: Translatology in the Digital Age (MoTra21)},
pages = {1-7},
publisher = {International Committee on Computational Linguistics},
address = {Iceland (Online)},
abstract = {Translationese data is a scarce and valuable resource. Traditionally, the proceedings of the European Parliament have been used for studying translationese phenomena since their metadata allows to distinguish between original and translated texts. However, translations are not always direct and we hypothesise that a pivot (also called ”relay”) language might alter the conclusions on translationese effects. In this work, we (i) isolate translations that have been done without an intermediate language in the Europarl proceedings from those that might have used a pivot language, and (ii) build comparable and parallel corpora with data aligned across multiple languages that therefore can be used for both machine translation and translation studies.},
pubstate = {published},
type = {inproceedings}
}

Copy BibTeX to Clipboard

Altering the content of an image with photo editing tools is a tedious task for an inexperienced user. Especially, when modifying the visual attributes of a specific object in an image without affecting other constituents such as background etc. To simplify the process of image manipulation and to provide more control to users, it is better to utilize a simpler interface like natural language. Therefore, in this paper, we address the challenge of manipulating images using natural language description. We propose the Two-sidEd Attentive conditional Generative Adversarial Network (TEA-cGAN) to generate semantically manipulated images while preserving other contents such as background intact. TEA-cGAN uses fine-grained attention both in the generator and discriminator of Generative Adversarial Network (GAN) based framework at different scales. Experimental results show that TEA-cGAN which generates 128×128 and 256×256 resolution images outperforms existing methods on CUB and Oxford-102 datasets both quantitatively and qualitatively.

• https://www.sciencedirect.com/science/article/pii/S0893608020303257
• https://doi.org/10.1016/j.neunet.2020.09.002

@article{zhumogadala:2020,
title = {Image manipulation with natural language using Two-sided Attentive Conditional Generative Adversarial Network},
author = {Dawei Zhu and Aditya Mogadala and Dietrich Klakow},
url = {https://www.sciencedirect.com/science/article/pii/S0893608020303257},
doi = {https://doi.org/10.1016/j.neunet.2020.09.002},
year = {2021},
date = {2021},
journal = {Neural Networks},
pages = {207-217},
volume = {136},
abstract = {Altering the content of an image with photo editing tools is a tedious task for an inexperienced user. Especially, when modifying the visual attributes of a specific object in an image without affecting other constituents such as background etc. To simplify the process of image manipulation and to provide more control to users, it is better to utilize a simpler interface like natural language. Therefore, in this paper, we address the challenge of manipulating images using natural language description. We propose the Two-sidEd Attentive conditional Generative Adversarial Network (TEA-cGAN) to generate semantically manipulated images while preserving other contents such as background intact. TEA-cGAN uses fine-grained attention both in the generator and discriminator of Generative Adversarial Network (GAN) based framework at different scales. Experimental results show that TEA-cGAN which generates 128x128 and 256x256 resolution images outperforms existing methods on CUB and Oxford-102 datasets both quantitatively and qualitatively.},
pubstate = {published},
type = {article}
}

Copy BibTeX to Clipboard

The interest in Artificial Intelligence (AI) and its applications has seen unprecedented growth in the last few years. This success can be partly attributed to the advancements made in the sub-fields of AI such as Machine Learning (ML), Computer Vision (CV), and Natural Language Processing (NLP). The largest of the growths in these fields has been made possible with deep learning, a sub-area of machine learning, which uses the principles of artificial neural networks. This has created significant interest in the integration of vision and language. The tasks are designed such that they perfectly embrace the ideas of deep learning. In this survey, we focus on ten prominent tasks that integrate language and vision by discussing their problem formulations, methods, existing datasets, evaluation measures, and compare the results obtained with corresponding state-of-the-art methods. Our efforts go beyond earlier surveys which are either task-specific or concentrate only on one type of visual content, i.e., image or video. Furthermore, we also provide some potential future directions in this field of research with an anticipation that this survey brings in innovative thoughts and ideas to address the existing challenges and build new applications.

• 1907.09358     (4.93MB)
• https://arxiv.org/abs/1907.09358
• https://doi.org/10.1613/jair.1.11688

@article{mogadala2021trends,
title = {Trends in Integration of Vision and Language Research: A Survey of Tasks, Datasets, and Methods},
author = {Aditya Mogadala and Marimuthu Kalimuthu and Dietrich Klakow},
url = {https://arxiv.org/abs/1907.09358},
doi = {https://doi.org/10.1613/jair.1.11688},
year = {2021},
date = {2021},
journal = {Journal of Artificial Intelligence Research},
pages = {1183-1317},
publisher = {Access Foundation},
volume = {71},
abstract = {The interest in Artificial Intelligence (AI) and its applications has seen unprecedented growth in the last few years. This success can be partly attributed to the advancements made in the sub-fields of AI such as Machine Learning (ML), Computer Vision (CV), and Natural Language Processing (NLP). The largest of the growths in these fields has been made possible with deep learning, a sub-area of machine learning, which uses the principles of artificial neural networks. This has created significant interest in the integration of vision and language. The tasks are designed such that they perfectly embrace the ideas of deep learning. In this survey, we focus on ten prominent tasks that integrate language and vision by discussing their problem formulations, methods, existing datasets, evaluation measures, and compare the results obtained with corresponding state-of-the-art methods. Our efforts go beyond earlier surveys which are either task-specific or concentrate only on one type of visual content, i.e., image or video. Furthermore, we also provide some potential future directions in this field of research with an anticipation that this survey brings in innovative thoughts and ideas to address the existing challenges and build new applications.},
pubstate = {published},
type = {article}
}

Copy BibTeX to Clipboard

Implicit discourse relation classification is a challenging task, in particular when the text domain is different from the standard Penn Discourse Treebank (PDTB; Prasad et al., 2008) training corpus domain (Wall Street Journal in 1990s). We here tackle the task of implicit discourse relation classification on the biomedical domain, for which the Biomedical Discourse Relation Bank (BioDRB; Prasad et al., 2011) is available. We show that entity information can be used to improve discourse relational argument representation. In a first step, we show that explicitly marked instances that are content-wise similar to the target relations can be used to achieve good performance in the cross-domain setting using a simple unsupervised voting pipeline. As a further step, we show that with the linked entity information from the first step, a transformer which is augmented with entity-related information (KBERT; Liu et al., 2020) sets the new state of the art performance on the dataset, outperforming the large pre-trained BioBERT (Lee et al., 2020) model by 2% points.

• https://aclanthology.org/2021.acl-short.116.pdf

@inproceedings{shi2021entity,
title = {Entity Enhancement for Implicit Discourse Relation Classification in the Biomedical Domain},
author = {Wei Shi and Vera Demberg},
url = {https://aclanthology.org/2021.acl-short.116.pdf},
year = {2021},
date = {2021},
booktitle = {Proceedings of the Joint Conference of the 59th Annual Meeting of theAssociation for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (ACL-IJCNLP 2021)},
abstract = {Implicit discourse relation classification is a challenging task, in particular when the text domain is different from the standard Penn Discourse Treebank (PDTB; Prasad et al., 2008) training corpus domain (Wall Street Journal in 1990s). We here tackle the task of implicit discourse relation classification on the biomedical domain, for which the Biomedical Discourse Relation Bank (BioDRB; Prasad et al., 2011) is available. We show that entity information can be used to improve discourse relational argument representation. In a first step, we show that explicitly marked instances that are content-wise similar to the target relations can be used to achieve good performance in the cross-domain setting using a simple unsupervised voting pipeline. As a further step, we show that with the linked entity information from the first step, a transformer which is augmented with entity-related information (KBERT; Liu et al., 2020) sets the new state of the art performance on the dataset, outperforming the large pre-trained BioBERT (Lee et al., 2020) model by 2% points.},
pubstate = {published},
type = {inproceedings}
}

Copy BibTeX to Clipboard

Cross-linguistic research on discourse structure and coherence marking requires discourse-annotated corpora and connective lexicons in a large number of languages. However, the availability of such resources is limited, especially for languages for which linguistic resources are scarce in general, such as Nigerian Pidgin. In this study, we demonstrate how a semi-automatic approach can be used to source connectives and their relation senses and develop a discourse-annotated corpus in a low-resource language. Connectives and their relation senses were extracted from a parallel corpus combining automatic (PDTB end-to-end parser) and manual annotations. This resulted in Naija-Lex, a lexicon of discourse connectives in Nigerian Pidgin with English translations. The lexicon shows that the majority of Nigerian Pidgin connectives are borrowed from its English lexifier, but that there are also some connectives that are unique to Nigerian Pidgin.

• 2021.codi-main.8     (0.36MB)
• https://aclanthology.org/2021.codi-main.8/
• https://doi.org/10.18653/v1/2021.codi-main.8

@inproceedings{marchal-etal-2021-semi,
title = {Semi-automatic discourse annotation in a low-resource language: Developing a connective lexicon for Nigerian Pidgin},
author = {Marian Marchal and Merel Scholman and Vera Demberg},
url = {https://aclanthology.org/2021.codi-main.8/},
doi = {https://doi.org/10.18653/v1/2021.codi-main.8},
year = {2021},
date = {2021},
booktitle = {Proceedings of the Second Workshop on Computational Approaches to Discourse (CODI 2021)},
pages = {84-94},
publisher = {Association for Computational Linguistics},
address = {Punta Cana, Dominican Republic and Online},
abstract = {Cross-linguistic research on discourse structure and coherence marking requires discourse-annotated corpora and connective lexicons in a large number of languages. However, the availability of such resources is limited, especially for languages for which linguistic resources are scarce in general, such as Nigerian Pidgin. In this study, we demonstrate how a semi-automatic approach can be used to source connectives and their relation senses and develop a discourse-annotated corpus in a low-resource language. Connectives and their relation senses were extracted from a parallel corpus combining automatic (PDTB end-to-end parser) and manual annotations. This resulted in Naija-Lex, a lexicon of discourse connectives in Nigerian Pidgin with English translations. The lexicon shows that the majority of Nigerian Pidgin connectives are borrowed from its English lexifier, but that there are also some connectives that are unique to Nigerian Pidgin.},
pubstate = {published},
type = {inproceedings}
}

Copy BibTeX to Clipboard