Papers by Drahomira Herrmannova
Beyond Pointwise Scores: Decomposed Criteria-Based Evaluation of LLM Responses (2025.emnlp-industry)
Copied to clipboard
| Challenge: | DeCE is model-agnostic and domain-general, requiring no predefined taxonomies or handcrafted rubrics. |
| Approach: | They propose a decomposed LLM evaluation framework that separates accuracy and recall from accuracy and relevance. |
| Outcome: | The proposed framework achieves stronger correlation with expert judgments than traditional metrics and pointwise LLM scoring. |
A Mixed-Method Design Approach for Empirically Based Selection of Unbiased Data Annotators (2021.findings-acl)
Copied to clipboard
| Challenge: | Current approaches to selecting annotators are limited at the policy-guidance level, rendering them unusable for machine learning practitioners. |
| Approach: | They propose a method that is functional, adaptable, and simpler to implement in selecting unbiased annotators for any machine learning problem. |
| Outcome: | The proposed approach is functional, adaptable, and simpler to implement on a real-world geopolitical problem. |
Analyzing Citation-Distance Networks for Evaluating Publication Impact (L18-1)
Copied to clipboard
| Challenge: | citation networks are used to study scholarly articles' semantic distances and their referencing patterns. |
| Approach: | They propose to analyze the semantic distance of scholarly articles in a citation network to uncover patterns that reflect scientific impact. |
| Outcome: | The proposed method combines semantic distance and content similarity to uncover scientific impact of articles in two different types of publications. |