Papers by Enrique Amigo
Evaluating Extreme Hierarchical Multi-label Classification (2022.acl-long)
Copied to clipboard
| Challenge: | Several natural language processing tasks are defined as a classification problem in its most complex form: Multi-label Hierarchical Extreme classification. |
| Approach: | They propose a classification metric inspired by the Information Contrast Model (ICM) they use a set of formal properties to analyze the evaluation metrics. |
| Outcome: | The proposed evaluation metrics are suitable for multi-label hierarchical extreme classification scenarios. |
On the Correspondence between the Squared Norm and Information Content in Text Embeddings (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing evidence of the correspondence between the squared norm of an embedding and the information content of the text it represents is lacking. |
| Approach: | They propose to derive two sufficient theoretical conditions for this correspondence to hold in embedding models. |
| Outcome: | The proposed embeddings exhibit a strong correspondence with the word embeddables and the subword token composition functions. |
An Effectiveness Metric for Ordinal Classification: Formal Properties and Experimental Results (2020.acl-main)
Copied to clipboard
| Challenge: | Existing Ordinal Classification metrics ignore the ordering between items or assume additional information. |
| Approach: | They propose a Closeness Evaluation Measure for Ordinal Classification based on Measurement Theory and Information Theory. |
| Outcome: | The proposed metric captures quality aspects from different traditional tasks simultaneously. |
Evaluating Sequence Labeling on the basis of Information Theory (2025.acl-long)
Copied to clipboard
| Challenge: | Existing metric families focus on certain aspects of sequence labeling tasks. |
| Approach: | They propose a metric that measures how much information each token contributes depending on different aspects of the sequence. |
| Outcome: | The proposed metric can satisfy all properties simultaneously. |
Bilingual Evaluation of Language Models on General Knowledge in University Entrance Exams with Minimal Contamination (2025.coling-main)
Copied to clipboard
Eva Sánchez Salido, Roser Morante, Julio Gonzalo, Guillermo Marco, Jorge Carrillo-de-Albornoz, Laura Plaza, Enrique Amigo, Andrés Fernandez García, Alejandro Benito-Santos, Adrián Ghajari Espinosa, Victor Fresno
| Challenge: | Existing benchmarks for Large Language Models have been proposed as single-task evaluations, but they are not fully comprehensive. |
| Approach: | They present a bilingual dataset that contains 1003 multiple-choice questions in Spanish and English. |
| Outcome: | The proposed model ranking is almost identical to the one obtained with MMLU . |