- ZeitschriftenartikelStance Detection Benchmark: How Robust is Your Stance Detection?(KI - Künstliche Intelligenz: Vol. 35, No. 0, 2021) Schiller, Benjamin; Daxenberger, Johannes; Gurevych, IrynaStance detection (StD) aims to detect an author’s stance towards a certain topic and has become a key component in applications like fake news detection, claim validation, or argument search. However, while stance is easily detected by humans, machine learning (ML) models are clearly falling short of this task. Given the major differences in dataset sizes and framing of StD (e.g. number of classes and inputs), ML models trained on a single dataset usually generalize poorly to other domains. Hence, we introduce a StD benchmark that allows to compare ML models against a wide variety of heterogeneous StD datasets to evaluate them for generalizability and robustness. Moreover, the framework is designed for easy integration of new datasets and probing methods for robustness. Amongst several baseline models, we define a model that learns from all ten StD datasets of various domains in a multi-dataset learning (MDL) setting and present new state-of-the-art results on five of the datasets. Yet, the models still perform well below human capabilities and even simple perturbations of the original test samples (adversarial attacks) severely hurt the performance of MDL models. Deeper investigation suggests overfitting on dataset biases as the main reason for the decreased robustness. Our analysis emphasizes the need of focus on robustness and de-biasing strategies in multi-task learning approaches. To foster research on this important topic, we release the dataset splits, code, and fine-tuned weights.
- ZeitschriftenartikelEmbodied Human Computer Interaction(KI - Künstliche Intelligenz: Vol. 35, No. 0, 2021) Pustejovsky, James; Krishnaswamy, NikhilIn this paper, we argue that embodiment can play an important role in the design and modeling of systems developed for Human Computer Interaction. To this end, we describe a simulation platform for building Embodied Human Computer Interactions (EHCI). This system, VoxWorld, enables multimodal dialogue systems that communicate through language, gesture, action, facial expressions, and gaze tracking, in the context of task-oriented interactions. A multimodal simulation is an embodied 3D virtual realization of both the situational environment and the co-situated agents, as well as the most salient content denoted by communicative acts in a discourse. It is built on the modeling language VoxML (Pustejovsky and Krishnaswamy in VoxML: a visualization modeling language, proceedings of LREC, 2016), which encodes objects with rich semantic typing and action affordances, and actions themselves as multimodal programs, enabling contextually salient inferences and decisions in the environment. VoxWorld enables an embodied HCI by situating both human and artificial agents within the same virtual simulation environment, where they share perceptual and epistemic common ground. We discuss the formal and computational underpinnings of embodiment and common ground, how they interact and specify parameters of the interaction between humans and artificial agents, and demonstrate behaviors and types of interactions on different classes of artificial agents.
- ZeitschriftenartikelConsciousness: Just Another Technique?(KI - Künstliche Intelligenz: Vol. 35, No. 0, 2021) Barthelmeß, Ulrike; Furbach, UlrichThis note is intended as a contribution to the discussion whether artificial systems can have consciousness. Based on the notion of Tononi’s Information Integration Theory we will argue, that AI systems that have to reason with large knowledge bases need such techniques in order to handle them efficiently. We furthermore discuss mind-wandering and creativity on this basis.
- ZeitschriftenartikelArtificial Intelligence: Mind, Computer and the Dance of the Wu Li Masters(KI - Künstliche Intelligenz: Vol. 35, No. 0, 2021) Siekmann, JörgIn these days of exuberant fantasies about the future development of artificial intelligence—mostly written by people who have never in their lives developed an AI program—the GFFT (Society for the Promotion of Technology Transfer) has also unleashed a competition on future AI scenarios to honour Wolfgang Bibel. Because I was allowed to give the laudatory speech for Wolfgang, I was also asked to contribute something to the pen. And because, despite everything else, it is not reprehensible to think about the future, I could not refrain from doing so. Here is my somewhat expanded contribution.
- ZeitschriftenartikelSpecial Issue on NLP & Semantics(KI - Künstliche Intelligenz: Vol. 35, No. 0, 2021) Hershcovich, Daniel; Donatelli, Lucia
- ZeitschriftenartikelNews(KI - Künstliche Intelligenz: Vol. 35, No. 0, 2021)
- ZeitschriftenartikelDissertation Abstract: The Syntax, Semantics and Pragmatics of Japanese Addressee-Honorific Markers(KI - Künstliche Intelligenz: Vol. 35, No. 0, 2021) Yamada, AkitakaThis dissertation is a case study of filling the gap between the two disciplines about human inference systems: theoretical linguistics and statistics. The main linguistic instance examined in this study is honorificity, in particular honorificity encoded by the Japanese addressee-honorific marker (AHM) -mas . Its linguistic properties and its effect on our inference are given a systematic explanation, in such a way that the traditions of statistics and theoretical linguistics are both maximally respected. For the morphosyntax, -mas is distributed in an unexpected position. It is proposed that this is due to an agreement. For the inference, the dynamicity is modeled as a Bayesian update, and the trigger of the update is the denotation amenable to the proposal of the previous linguistic literature.
- ZeitschriftenartikelDesigning a Uniform Meaning Representation for Natural Language Processing(KI - Künstliche Intelligenz: Vol. 35, No. 0, 2021) Van Gysel, Jens E. L.; Vigus, Meagan; Chun, Jayeol; Lai, Kenneth; Moeller, Sarah; Yao, Jiarui; O’Gorman, Tim; Cowell, Andrew; Croft, William; Huang, Chu-Ren; Hajič, Jan; Martin, James H.; Oepen, Stephan; Palmer, Martha; Pustejovsky, James; Vallejos, Rosa; Xue, NianwenIn this paper we present Uniform Meaning Representation (UMR), a meaning representation designed to annotate the semantic content of a text. UMR is primarily based on Abstract Meaning Representation (AMR), an annotation framework initially designed for English, but also draws from other meaning representations. UMR extends AMR to other languages, particularly morphologically complex, low-resource languages. UMR also adds features to AMR that are critical to semantic interpretation and enhances AMR by proposing a companion document-level representation that captures linguistic phenomena such as coreference as well as temporal and modal dependencies that potentially go beyond sentence boundaries.
- ZeitschriftenartikelDissertation Abstract:Learning High Precision Lexical Inferences(KI - Künstliche Intelligenz: Vol. 35, No. 0, 2021) Shwartz, VeredThe fundamental goal of natural language processing is to build models capable of human-level understanding of natural language. One of the obstacles to building such models is lexical variability , i.e. the ability to express the same meaning in various ways. Existing text representations excel at capturing relatedness (e.g. blue / red ), but they lack the fine-grained distinction of the specific semantic relation between a pair of words. This article is a summary of a Ph.D. dissertation submitted to Bar-Ilan University in 2019, under the supervision of Professor Ido Dagan of the Computer Science Department. The dissertation explored methods for recognizing and extracting semantic relationships between concepts ( cat is a type of animal ), the constituents of noun compounds (baby oil is oil for babies), and verbal phrases (‘X died at Y’ means the same as ‘X lived until Y’ in certain contexts). The proposed models outperform highly competitive baselines and improve the state-of-the-art in several benchmarks. The dissertation concludes in discussing two challenges in the way of human-level language understanding: developing more accurate text representations and learning to read between the lines.
- ZeitschriftenartikelThe PANDA Framework for Hierarchical Planning(KI - Künstliche Intelligenz: Vol. 35, No. 0, 2021) Höller, Daniel; Behnke, Gregor; Bercher, Pascal; Biundo, SusanneDuring the last years, much progress has been made in hierarchical planning towards domain-independent systems that come with sophisticated techniques to solve planning problems instead of relying on advice in the input model. Several of these novel methods have been integrated into the PANDA framework , which is a software system to reason about hierarchical planning tasks. Besides solvers for planning problems based on plan space search, progression search, and translation to propositional logic, it also includes techniques for related problems like plan repair, plan and goal recognition, or plan verification. These various techniques share a common infrastructure, like e.g. a standard input language or components for grounding and reachability analysis. This article gives an overview over the PANDA framework, introduces the basic techniques from a high level perspective, and surveys the literature describing the diverse components in detail.