Natalia Umansky
Biographic Data
| ID | 4383708 |
|---|---|
| NAME | Natalia Umansky |
| GIVEN NAMES | Natalia |
| FAMILY NAME | Umansky |
| SIGNATURE | UMANSKY N |
| AFFILIATIONS | University College Dublin |
| ORCID | 0000-0002-8059-5328 |
| VERIFIED | Yes |
| TOTAL WORKS | 3 |
| TOTAL CITATIONS | 2 |
| WORKS AS AUTHOR | 3 |
| WORKS AS EDITOR | 0 |
| FIRST PUBLICATION YEAR | 2022 |
| LATEST PUBLICATION YEAR | 2026 |
| H-INDEX | 1 |
Improving hate speech detection with large language models
Efforts to curb online hate speech depend on our ability to reliably detect it at scale. Previous studies have highlighted the strong zero-shot classification performance of large language models (LLMs), offering a potential tool to efficiently identify harmful content. Yet for complex and ambivalent tasks like hate speech detection, pre-trained LLMs can be insufficient and carry systemic biases. Domain-specific models fine-tuned for the given ta…
Speaking in unison? Explaining the role of agenda-setter constellations in the ECB policy agenda using a network-based approach
Policy agendas are a well-studied institutional level phenomenon that capture the set of policy issues that an institution pays attention to over time. They are emergent in nature in that individual behaviour shapes institutional level outcomes when policy makers allocate attention to policy issues. To examine the link between individual-level actions and system-level outcomes we introduce the concept of the agenda-setting constellation, defined …
Who gets a say in this? Speaking security on social media
Can social media revert the top-down dynamics of securitization? Limited by the notion that ‘security is only articulated in an institutional voice by the elites’, the role of non-elite actors has remained understudied. Only recently has it been proposed that lay actors can become influential security agents through their online activity. However, social media’s capacity to revert the top-down dynamics of securitization remains contended. To expl…
Speaking in unison? Explaining the role of agenda-setter constellations in the ECB policy agenda using a network-based approach
Policy agendas are a well-studied institutional level phenomenon that capture the set of policy issues that an institution pays attention to over time. They are emergent in nature in that individual behaviour shapes institutional level outcomes when policy makers allocate attention to policy issues. To examine the link between individual-level actions and system-level outcomes we introduce the concept of the agenda-setting constellation, defined …
Who gets a say in this? Speaking security on social media
Can social media revert the top-down dynamics of securitization? Limited by the notion that ‘security is only articulated in an institutional voice by the elites’, the role of non-elite actors has remained understudied. Only recently has it been proposed that lay actors can become influential security agents through their online activity. However, social media’s capacity to revert the top-down dynamics of securitization remains contended. To expl…
Who gets a say in this? Speaking security on social media
Can social media revert the top-down dynamics of securitization? Limited by the notion that ‘security is only articulated in an institutional voice by the elites’, the role of non-elite actors has remained understudied. Only recently has it been proposed that lay actors can become influential security agents through their online activity. However, social media’s capacity to revert the top-down dynamics of securitization remains contended. To expl…
Speaking in unison? Explaining the role of agenda-setter constellations in the ECB policy agenda using a network-based approach
Policy agendas are a well-studied institutional level phenomenon that capture the set of policy issues that an institution pays attention to over time. They are emergent in nature in that individual behaviour shapes institutional level outcomes when policy makers allocate attention to policy issues. To examine the link between individual-level actions and system-level outcomes we introduce the concept of the agenda-setting constellation, defined …
Improving hate speech detection with large language models
Efforts to curb online hate speech depend on our ability to reliably detect it at scale. Previous studies have highlighted the strong zero-shot classification performance of large language models (LLMs), offering a potential tool to efficiently identify harmful content. Yet for complex and ambivalent tasks like hate speech detection, pre-trained LLMs can be insufficient and carry systemic biases. Domain-specific models fine-tuned for the given ta…
Computer Science (2 works) · Context (archaeology (2 works) · Economics (2 works) · Law (2 works) · Political science (2 works) · Annotation (1 work) · Constellation (1 work) · Crowdsourcing (1 work) · Electoral Systems and Political Participation (1 work) · Elite (1 work)