Cennet Oguz's research while affiliated with Deutsches Forschungszentrum für Künstliche Intelligenz and other places

Publications (9)

Preprint
Full-text available
"How can we animate 3D-characters from a movie script or move robots by simply telling them what we would like them to do?" "How unstructured and complex can we make a sentence and still generate plausible movements from it?" These are questions that need to be answered in the long-run, as the field is still in its infancy. Inspired by these proble...
Preprint
Full-text available
Metric-based learning is a well-known family of methods for few-shot learning, especially in computer vision. Recently, they have been used in many natural language processing applications but not for slot tagging. In this paper, we explore metric-based learning methods in the slot tagging task and propose a novel metric-based learning architecture...
Preprint
Full-text available
We simulate first- and second-order context overlap and show that Skip-Gram with Negative Sampling is similar to Singular Value Decomposition in capturing second-order co-occurrence information, while Pointwise Mutual Information is agnostic to it. We support the results with an empirical study finding that the models react differently when provide...

Citations

... The generation of realistic 3D human motions has applications in virtual reality, the games industry, and any application that requires motion capture data. Recently, controlling 3D human motion synthesis with semantics has received increasing attention [13,30,11,31]. The task concerns inputting semantics in the form of categorical actions, or free-form natural language descriptions, and outputting a series of 3D body poses. ...
... In this section, we analyze the results of the four teams that participated in the anaphora resolution track and submitted a shared task paper, namely the team from Emory University (Xu and Choi, 2021) (henceforth Emory), the team from the University of Texas at Dallas (Kobayashi et al., 2021) (henceforth UTD), the team from Korea University (Kim et al., 2021) (henceforth KU), and the DFKI team (Anikina et al., 2021) ...
... Owing to the need to handle myriad user expressions, the dialogue understanding module in taskoriented dialogue systems faces particularly challenges from data deficiencies [25]. To remedy this, the two main tasks of DU: intent detection [28,37,42,72] and slot tagging [26,45,60], have been widely studied in the few-shot setting. However, previous works mainly address only individual tasks or the two tasks separately [70]. ...
... Slot filling, as an important part of the task-oriented dialogue system, is mainly used to extract specific information in user utterances. Traditional supervised methods have shown remarkable performance in slot filling tasks (Liu and Lane, 2016;Goo et al., 2018;E et al., 2019;He et al., 2020b;Wu et al., 2020;He et al., 2020a;Oguz and Vu, 2020;Qin et al., 2020), but they require a large amount of domain-specific labeled data. ...
... Indeed, empirical tests of SGNS and PMI demonstrate that SGNS is extremely capable of preserving second-order context overlap-even weighting this higher than first-order context overlap-while PMI is completely incapable of capturing it at all. In a simulation experiment performed by Schlechtweg et al. 34 , the average cosine distance between words with first-and second-order context overlap were 0.11 and 0.00 respectively using SGNS and 0.51 and 1.0 using PMI. While matrix factorization of the PMI matrix is also able to capture such higher-order effects, Levy and Goldberg establish that in practice SGNS arrives at a different result than factorization of the PMI matrix, and that pure factorization does not perform well on many NLP tasks 33 . ...