Papers by Katharina Reinecke
NLPositionality: Characterizing Design Biases of Datasets and Models (2023.acl-long)
Copied to clipboard
| Challenge: | Design biases in NLP systems often stem from creator’s positionality, i.e., views and lived experiences shaped by identity and background. |
| Approach: | They propose a framework for characterizing design biases and quantifying the positionality of NLP datasets and models. |
| Outcome: | The proposed framework characterizes design biases and quantifies alignment with dataset labels and model predictions. |
Biased LLMs can Influence Political Decision-Making (2025.acl-long)
Copied to clipboard
Jillian Fisher, Shangbin Feng, Robert Aron, Thomas Richardson, Yejin Choi, Daniel W Fisher, Jennifer Pan, Yulia Tsvetkov, Katharina Reinecke
| Challenge: | Recent studies have found that biased LLMs can influence decisions in areas such as medical classifications and educational hiring. |
| Approach: | They conducted two interactive experiments on partisan bias in large language models while completing tasks with either a biased liberal, biased conservative, or unbiased control model. |
| Outcome: | The results show that prior knowledge of AI is weakly correlated with a reduction of the bias, suggesting that AI education can be crucial for mitigating bias effects. |
Generating Scientific Definitions with Controllable Complexity (2022.acl-long)
Copied to clipboard
| Challenge: | Unfamiliar terminology and complex language can make understanding science difficult for readers. |
| Approach: | They propose a task and dataset for defining scientific terms and controlling the complexity of generated definitions by a sequence-to-sequence approach. |
| Outcome: | The proposed system is based on a sequence-to-sequence approach and human evaluations show it offers superior fluency while controlling complexity. |
Writing Strategies for Science Communication: Data and Computational Analysis (2020.emnlp-main)
Copied to clipboard
| Challenge: | Existing science communication guides do not provide empirical evidence for how their strategies are used in practice. |
| Approach: | They propose to use prescriptive writing strategies to identify and train human-readable annotations that can be automatically recognized by a corpus of 128k science writing documents in English. |
| Outcome: | The proposed system can be used to detect and suggest writing strategies for scientists by allowing them to automatically recognize them. |
NormAd: A Framework for Measuring the Cultural Adaptability of Large Language Models (2025.naacl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) are widely used and engage millions of users from diverse contexts and cultures. |
| Approach: | They propose an evaluation framework to assess LLMs’ cultural adaptability by measuring their ability to judge social acceptability across varying levels of cultural norm specificity. |
| Outcome: | The proposed model shows stronger adaptability to English-centric cultures over those from the Global South. |
Gendered Mental Health Stigma in Masked Language Models (2022.emnlp-main)
Copied to clipboard
Inna Lin, Lucille Njoo, Anjalie Field, Ashish Sharma, Katharina Reinecke, Tim Althoff, Yulia Tsvetkov
| Challenge: | Mental health stigma prevents many individuals from receiving appropriate care, and social psychology studies have shown that mental health tends to be overlooked in men. |
| Approach: | They propose to use clinical psychology literature to curate prompts, then evaluate models’ propensity to generate gendered words. |
| Outcome: | The proposed framework captures stigma about gender in mental health and is more likely to predict female subjects than male in sentences about mental health conditions (32% vs. 19%), and this disparity is exacerbated for sentences that indicate treatment-seeking behavior. |