A5069780274- Natural Language Processing Techniques
- Topic Modeling
- Semantic Web and Ontologies
- Text Readability and Simplification
- Syntax, Semantics, Linguistic Variation
- Computational and Text Analysis Methods
- Lexicography and Language Studies
- Speech Recognition and Synthesis
- Handwritten Text Recognition Techniques
- Language, Linguistics, Cultural Analysis
- Speech and dialogue systems
- Advanced Text Analysis Techniques
- Advanced Algebra and Logic
University of Colorado Boulder
2021-2023
University of Colorado System
2019-2023
Xinya Du, Zixuan Zhang, Sha Li, Pengfei Yu, Hongwei Wang, Tuan Lai, Xudong Lin, Ziqi Iris Liu, Ben Zhou, Haoyang Wen, Manling Darryl Hannan, Jie Lei, Hyounghun Kim, Rotem Dror, Haoyu Michael Regan, Qi Zeng, Qing Lyu, Charles Carl Edwards, Xiaomeng Jin, Yizhu Jiao, Ghazaleh Kazeminejad, Zhenhailong Chris Callison-Burch, Mohit Bansal, Vondrick, Jiawei Han, Dan Roth, Shih-Fu Chang, Martha Palmer, Heng Ji. Proceedings of the 2022 Conference North American Chapter Association for Computational...
The need for deeper semantic processing of human language by our natural systems is evidenced their still-unreliable performance on inferencing tasks, even using deep learning techniques. These tasks require the detection subtle interactions between participants in events, sequencing subevents that are often not explicitly mentioned, and changes to various across an event. Human beings can perform this when sparse lexical items involved, suggesting linguistic insights into these abilities...
This paper discusses the challenges in creating pedagogical and research resources for Arapaho language.Because of complex morphology this language, printed cannot provide typical useful functions that they can many other languages.We discuss faced by most learners researchers working with polysynthetic languages, which is an excellent example, as well some currently implemented solutions computer including a lexical database, morphological parser, concordancer.The construction finite-state...
Piyush Mishra, Akanksha Malhotra, Susan Windisch Brown, Martha Palmer, Ghazaleh Kazeminejad. Proceedings of the 59th Annual Meeting Association for Computational Linguistics and 11th International Joint Conference on Natural Language Processing: System Demonstrations. 2021.
Computational lexical resources such as WordNet, PropBank, VerbNet, and FrameNet are in regular use various NLP applications, assisting the never-ending quest for richer, more precise semantic representations. Coherent class-based organization of units VerbNet can improve efficiency processing by clustering similar items together sharing descriptions. However, class members sometimes quite different, both gloss over useful fine-grained distinctions. officially eschews syntactic...
Tracking entity states is a natural language processing task assumed to require human annotation. In order reduce the time and expenses associated with annotation, we introduce new method automatically extract states, including location existence state of entities, following Dalvi et al. (2018) Tandon (2020). For this purpose, rely primarily on semantic representations generated by art VerbNet parser (Gung, 2020), entities (event participants) their based predicates representation, which in...
Neural encoder-decoder models are usually applied to morphology learning as an end-to-end process without considering the underlying phonological representations that linguists posit abstract forms before morphophonological rules applied. Finite State Transducers for morphology, on other hand, developed contain these intermediate representation. This paper shows training a bidirectional two-step model of Arapaho verbs learn two separate mappings between tags and morphemes surface allomorphs...
The task of entity state tracking aims to automatically analyze procedural texts – that describe a step-by-step process (e.g. baking recipe). Specifically, the goal is track various states entities participating in given process. Some challenges for this NLP include annotated data scarcity and annotators’ reliance on commonsense knowledge annotate implicit information. Zhang et al. (2021) successfully incorporated entity-centric from ConceptNet into their BERT-based neural-symbolic...
Semantic role labeling (SRL) has multiple disjoint label sets, e.g., VerbNet and PropBank. Creating these datasets is challenging, therefore a natural question how to use each one help the other. Prior work shown that cross-task interaction helps, but only explored multitask learning so far. A common issue with multi-task setup argument sequences are still separately decoded, running risk of generating structurally inconsistent (as per lexicons like Semlink). In this paper, we eliminate such...
Semantic role labeling (SRL) has multiple disjoint label sets, e.g., VerbNet and PropBank. Creating these datasets is challenging, therefore a natural question how to use each one help the other. Prior work shown that cross-task interaction helps, but only explored multitask learning so far. A common issue with multi-task setup argument sequences are still separately decoded, running risk of generating structurally inconsistent (as per lexicons like Semlink). In this paper, we eliminate such...