A5103060010- Natural Language Processing Techniques
- Topic Modeling
- earthquake and tectonic studies
- Speech and dialogue systems
- Text Readability and Simplification
- AI in Service Interactions
- Geology and Paleoclimatology Research
- Geological formations and processes
- Social Media and Politics
- Misinformation and Its Impacts
- Geological and Geophysical Studies
- Seismic Waves and Analysis
- Geological and Geochemical Analysis
- Second Language Acquisition and Learning
- Hate Speech and Cyberbullying Detection
- Landslides and related hazards
- Linguistic Studies and Language Acquisition
- Marine and fisheries research
- Bullying, Victimization, and Aggression
- Artificial Intelligence in Games
- Tunneling and Rock Mechanics
- History of Medical Practice
- Coastal and Marine Management
- Sentiment Analysis and Opinion Mining
- Music Technology and Sound Studies
National Institute of Water and Atmospheric Research
2022-2025
University of California, Davis
2019-2023
University of Canterbury
2020-2022
Institute for Computational Linguistics “A. Zampolli”
2020
University of Pisa
2020
Center for Applied Linguistics
2020
Anthropogenic impacts are increasingly affecting deep-sea environments, including seafloor sediment disturbances by bottom trawling and mining. Fieldwork in the 'Resilience Of Benthic Ecosystems to Sedimentation' (ROBES) project were conducted 2018–2020 on 400 m-deep Chatham Rise crest, eastern Aotearoa New Zealand. Water column turbidity data, traps near-seabed moorings benthic landers surficial sediments from multi-corers provided baseline post-impact information following an artificially...
Most scholars focus on the prevalence and democratic effects of (partisan) news exposure. This misses large parts online activities a majority politically disinterested citizens. Although political content also appears outside outlets may profoundly shape public opinion, its are under-studied at scale. project combines three-wave panel survey data from three countries (total N = 7,266) with behavioral same participants (over 106M visits). We create multi-lingual classifier to identify both...
Incivility in social media has become a major concern of the public, who perceive uncivil online interactions to be both widespread and increasing. This study provides descriptive account incivility dynamics over past 11 years by examining trends three main categories interactions: political, mixed, non-political. Using longitudinal data from Reddit that accounts for 95% entire universe across relying on combination supervised machine learning models traditional statistical inference, found...
ABSTRACT Subduction trenches receive sediment from gravity flows sourced transverse pathways and trench parallel axial transport pathways. Understanding the interplay between in shaping stratigraphic architectures is hindered by episodic nature of sedimentary limited datasets, yet such insights are crucial for reconstructing flow interpreting records. We investigate routing to northern Hikurangi Trough New Zealand using a combination multibeam, 2D 3D seismic reflection International Ocean...
Abstract Documenting and characterizing past submarine landslides is fundamental to understanding their distribution frequency through time, critical assessing the associated hazard. The widespread availability of marine geophysical data at active Hikurangi subduction margin, east Aotearoa New Zealand, provides an excellent basis map regional trends in landslide occurrence. We present a database that documents mass transport deposits (MTDs) 30 surveys, encompassing ∼45,400 line‐km 2D seismic...
Incivility is not only prevalent on online social media platforms, but also has concrete effects individual users, groups, and the platforms themselves. Given prevalence of incivility, challenges involved in human-based incivility detection, it urgent to develop validated versatile automatic approaches identifying uncivil posts comments. This project advances both a neural, BERT-based classifier as well logistic regression identify The trained dataset Reddit posts, which are annotated for...
Gunrock 2.0 is built on top of with an emphasis user adaptation. combines various neural natural language understanding modules, including named entity detection, linking, and dialog act prediction, to improve understanding. Its management a hierarchical model that handles topics, such as movies, music, sports. The system-level manager can handle question acknowledgment, error handling, additional functions, making downstream modules much easier design implement. also adapts its topic...
Dian Yu, Michelle Cohn, Yi Mang Yang, Chun Yen Chen, Weiming Wen, Jiaping Zhang, Mingyang Zhou, Kevin Jesse, Austin Chau, Antara Bhowmick, Shreenath Iyer, Giritheja Sreenivasulu, Sam Davidson, Ashwin Bhandare, Zhou Yu. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint (EMNLP-IJCNLP): System Demonstrations. 2019.
Kai-Hui Liang, Sam Davidson, Xun Yuan, Shehan Panditharatne, Chun-Yen Chen, Ryan Shea, Derek Pham, Yinghua Tan, Erik Voss, Luke Fryer. Proceedings of the 18th Workshop on Innovative Use NLP for Building Educational Applications (BEA 2023). 2023.
This paper reports on progress towards building an online language learning tool to provide learners with conversational experience by using dialog systems as conversation practice partners. Our system can adapt users' proficiency the fly. We also automatic grammar error feedback help users learn from their mistakes. According our first adopters, is entertaining and useful. Furthermore, we will technology community a large-scale dataset correction. next step make more adaptive user profile...
Abstract The initial stages of seamount subduction and associated deformation in an overriding accretionary wedge is rarely documented. Initial Bennett Knoll faulting the overlying strata along Hikurangi margin, New Zealand, are here studied using multibeam swath bathymetry, subbottom profiles, regional seismic reflection lines. Our results provide new insights into earliest collision at sediment-rich margins. Differential shortening front induced by initially accommodated conjugate...
This paper presents the Corpus of Written Spanish L2 and Heritage Speakers (COWS-L2H), a large corpus compositions written by North American university students learning Spanish. The goals this work are to (1) build learner writing that provides samples data from learners in context university, (2) contribute collected not only second language (L2) but also as heritage (SHL), (3) develop one few corpora provide longitudinal data.
Sam Davidson, Dian Yu, Zhou Yu. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint (EMNLP-IJCNLP). 2019.
Gunrock is the winner of 2018 Amazon Alexa Prize, as evaluated by coherence and engagement from both real users Amazon-selected expert conversationalists. We focus on understanding complex sentences having in-depth conversations in open domains. In this paper, we introduce some innovative system designs related validation analysis. Overall, found that produce longer to Gunrock, which are directly users' (e.g., ratings, number turns). Additionally, backstory queries about positively...
Alessio Miaschi, Sam Davidson, Dominique Brunato, Felice Dell'Orletta, Kenji Sagae, Claudia Helena Sanchez-Gutierrez, Giulia Venturi. Proceedings of the Fifteenth Workshop on Innovative Use NLP for Building Educational Applications. 2020.
One of the major impediments to development new task-oriented dialogue (TOD) systems is need for human evaluation at multiple stages and iterations process. In an effort move toward automated TOD, we propose a novel user simulator built using recently developed large pretrained language models (LLMs). order increase linguistic diversity our system relative related previous work, do not fine-tune LLMs used by on existing TOD datasets; rather use in-context learning prompt generate robust...
Although significant progress has been made in developing methods for Grammatical Error Correction (GEC), addressing word choice improvements notably lacking and enhancing sentence expressivity by replacing phrases with advanced expressions is an understudied aspect. In this paper, we focus on area present our investigation into the task of incorporating usage idiomatic student writing. To facilitate study, curate extensive training sets expert-annotated testing using real-world data...