Aida Kostikova
2024
Fine-Grained Detection of Solidarity for Women and Migrants in 155 Years of German Parliamentary Debates
Aida Kostikova
|
Dominik Beese
|
Benjamin Paassen
|
Ole Pütz
|
Gregor Wiedemann
|
Steffen Eger
Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
Solidarity is a crucial concept to understand social relations in societies. In this study, we investigate the frequency of (anti-)solidarity towards women and migrants in German parliamentary debates between 1867 and 2022. Using 2,864 manually annotated text snippets, we evaluate large language models (LLMs) like Llama 3, GPT-3.5, and GPT-4. We find that GPT-4 outperforms other models, approaching human annotation accuracy. Using GPT-4, we automatically annotate 18,300 further instances and find that solidarity with migrants outweighs anti-solidarity but that frequencies and solidarity types shift over time. Most importantly, group-based notions of (anti-)solidarity fade in favor of compassionate solidarity, focusing on the vulnerability of migrant groups, and exchange-based anti-solidarity, focusing on the lack of (economic) contribution. This study highlights the interplay of historical events, socio-economic needs, and political ideologies in shaping migration discourse and social cohesion.
2023
How adaptive is adaptive machine translation, really? A gender-neutral language use case
Aida Kostikova
|
Joke Daems
|
Todor Lazarov
Proceedings of the First Workshop on Gender-Inclusive Translation Technologies
This study examines the effectiveness of adaptive machine translation (AMT) for gender-neutral language (GNL) use in English-German translation using the ModernMT engine. It investigates gender bias in initial output and adaptability to two distinct GNL strategies, as well as the influence of translation memory (TM) use on adaptivity. Findings indicate that despite inherent gender bias, machine translation (MT) systems show potential for adapting to GNL with appropriate exposure and training, highlighting the importance of customisation, exposure to diverse examples, and better representation of different forms for enhancing gender-fair translation strategies.
Search
Co-authors
- Dominik Beese 1
- Benjamin Paassen 1
- Ole Pütz 1
- Gregor Wiedemann 1
- Steffen Eger 1
- show all...