Papers by Cecilia Ovesdotter Alm
Transfer Learning Methods for Domain Adaptation in Technical Logbook Datasets (2022.lrec-1)
Copied to clipboard
| Challenge: | Technical logbook data typically has both a domain, the field it comes from, and an application, what it is used for. |
| Approach: | They propose to use domain-specific technical language to identify technical logbook entries by using transfer learning to learn from different domains and from different datasets. |
| Outcome: | The proposed approach improves performance in all cases but one of the three domains studied. |
Unpacking the Interdependent Systems of Discrimination: Ableist Bias in NLP Systems through an Intersectional Lens (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Statistically significant results demonstrate that people with disabilities can be disadvantaged. |
| Approach: | They used a large-scale BERT language model to predict word predictions and found that people with disabilities can be disadvantaged. |
| Outcome: | The results show that people with disabilities can be disadvantaged and that gender and race identities can be discriminated against. |
Handling Extreme Class Imbalance in Technical Logbook Datasets (2021.acl-long)
Copied to clipboard
| Challenge: | Technical logbooks are a challenging and under-explored text type in automated event identification. |
| Approach: | They propose a feedback strategy that resamples the training data based on its error in the prediction process. |
| Outcome: | The proposed approach provides the best results for four different neural network models trained across a suite of technical logbook datasets from distinct technical domains. |
MULTICOLLAB: A Multimodal Corpus of Dialogues for Analyzing Collaboration and Frustration in Language (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing methods to study complex emotions when a speaker collaborates with a partner are limited. |
| Approach: | They propose to fuse a multimodal dialogue resource with transcribed speech and eye gaze data to create a highly multimodal corpus. |
| Outcome: | The proposed model improves classification accuracy by 21% over baseline using sensor and speech data in 4.5 seconds. |
FUSE - FrUstration and Surprise Expressions: A Subtle Emotional Multimodal Language Corpus (2024.lrec-main)
Copied to clipboard
| Challenge: | elicitation of frustration and surprise are understudied for emotion modeling in language, but are difficult to characterize. |
| Approach: | They propose a multimodal corpus for expressive task-based spoken language and dialogue focused on frustration and surprise, which are understudied for emotion modeling in language. |
| Outcome: | The proposed corpus provides both individual and dyadic multimodally grounded language. |