Papers by Cristina Sarasua
DREAM: Deployment of Recombination and Ensembles in Argument Mining (2023.emnlp-main)
Copied to clipboard
| Challenge: | Current approaches to Argument Mining (AM) take a holistic view of the overall pipeline. |
| Approach: | They propose a framework that allows for the (automated) combination of AM components instead of all-new solutions. |
| Outcome: | The proposed framework outperforms the best single systems in terms of accuracy measured by an AM benchmark. |
HypER: Literature-grounded Hypothesis Generation and Distillation with Provenance (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches focus on retrieval augmentation and focus on the quality of the output . Existing methods focus on generating a highly specific declarative statement ignoring the underlying reasoning process behind ideation. |
| Approach: | They propose a large language model that generates evidence-based hypotheses using literature-guided reasoning and a multi-task setting. |
| Outcome: | The proposed model outperforms the base model and generates evidence-grounded hypotheses with high feasibility and impact as judged by human experts. |