add new publications

content/en/publication/agirre-2024-ikergaitu/index.md

@@ -67,7 +67,7 @@ tags:
 - Basque
 
 # Display this page in a list of Featured pages?
-featured: true
+featured: false
 
 # Links
 url_pdf: https://www.ixa.eus/sites/default/files/dokumentuak/14014/2024_SEPLN___CEDI___iker_gaitu-4.pdf

content/en/publication/bengoetxea-2024-hitzvardial/cite.bib

@@ -0,0 +1,13 @@
+@article{bengoetxea2024hitzvardial,
+ abstract = {In this paper we present our submission for the NorSID Shared Task as part of the 2025 VarDial Workshop (Scherrer et al., 2025), consisting of three tasks: Intent Detection, Slot Filling and Dialect Identification, evaluated using data in different dialects of the Norwegian language. For Intent Detection and Slot Filling, we have fine-tuned a multitask model in a cross-lingual setting, to leverage the xSID dataset available in 17 languages. In the case of Dialect Identification, our final submission consists of a model fine-tuned on the provided development set, which has obtained the highest scores within our experiments. Our final results on the test set show that our models do not drop in performance compared to the development set, likely due to the domain-specificity of the dataset and the similar distribution of both subsets. Finally, we also report an in-depth analysis of the provided datasets and their artifacts, as well as other sets of experiments that have been carried out but did not yield the best results. Additionally, we present an analysis on the reasons why some methods have been more successful than others; mainly the impact of the combination of languages and domain-specificity of the training data on the results.},
+ archiveprefix = {arXiv},
+ author = {Jaione Bengoetxea and Mikel Zubillaga and Ekhi Azurmendi and Maite Heredia and Julen Etxaniz and Markel Ferro and Jeremy Barnes},
+ booktitle = {COLING 2025},
+ date = {2024-12-13},
+ eprint = {2412.10095},
+ keywords = {Natural Language Processing, Large Language Models, Deep Learning, Multilinguality, Dialects, Norwegian},
+ primaryclass = {cs.CL},
+ title = {HiTZ at VarDial 2025 NorSID: Overcoming Data Scarcity with Language Transfer and Automatic Data Annotation},
+ url = {https://arxiv.org/abs/2412.10095},
+ year = {2024}
+}

content/en/publication/bengoetxea-2024-hitzvardial/index.md

@@ -0,0 +1,100 @@
+---
+title: 'HiTZ at VarDial 2025 NorSID: Overcoming Data Scarcity with Language Transfer
+  and Automatic Data Annotation'
+
+# Authors
+# A YAML list of author names
+# If you created a profile for a user (e.g. the default `admin` user at `content/authors/admin/`), 
+# write the username (folder name) here, and it will be replaced with their full name and linked to their profile.
+authors:
+- Jaione Bengoetxea
+- Mikel Zubillaga
+- Ekhi Azurmendi
+- Maite Heredia
+- Julen Etxaniz
+- Markel Ferro
+- Jeremy Barnes
+
+# Author notes (such as 'Equal Contribution')
+# A YAML list of notes for each author in the above `authors` list
+author_notes: []
+
+date: '2024-12-13'
+
+# Date to publish webpage (NOT necessarily Bibtex publication's date).
+publishDate: '2025-01-03T18:58:39.112301Z'
+
+# Publication type.
+# A single CSL publication type but formatted as a YAML list (for Hugo requirements).
+publication_types:
+- article-journal
+
+# Publication name and optional abbreviated publication name.
+publication: '*COLING 2025*'
+publication_short: ''
+
+doi: ''
+
+abstract: 'In this paper we present our submission for the NorSID Shared Task as part
+  of the 2025 VarDial Workshop (Scherrer et al., 2025), consisting of three tasks:
+  Intent Detection, Slot Filling and Dialect Identification, evaluated using data
+  in different dialects of the Norwegian language. For Intent Detection and Slot Filling,
+  we have fine-tuned a multitask model in a cross-lingual setting, to leverage the
+  xSID dataset available in 17 languages. In the case of Dialect Identification, our
+  final submission consists of a model fine-tuned on the provided development set,
+  which has obtained the highest scores within our experiments. Our final results
+  on the test set show that our models do not drop in performance compared to the
+  development set, likely due to the domain-specificity of the dataset and the similar
+  distribution of both subsets. Finally, we also report an in-depth analysis of the
+  provided datasets and their artifacts, as well as other sets of experiments that
+  have been carried out but did not yield the best results. Additionally, we present
+  an analysis on the reasons why some methods have been more successful than others;
+  mainly the impact of the combination of languages and domain-specificity of the
+  training data on the results.'
+
+# Summary. An optional shortened abstract.
+summary: ''
+
+tags:
+- Natural Language Processing
+- Large Language Models
+- Deep Learning
+- Multilinguality
+- Dialects
+- Norwegian
+
+# Display this page in a list of Featured pages?
+featured: true
+
+# Links
+url_pdf: 'https://arxiv.org/pdf/2412.10095'
+url_code: 'https://github.com/hitz-zentroa/vardial-2025'
+url_dataset: ''
+url_poster: ''
+url_project: ''
+url_slides: ''
+url_source: ''
+url_video: ''
+
+# Custom links (uncomment lines below)
+# links:
+# - name: Custom Link
+#   url: http://example.org
+
+# Publication image
+# Add an image named `featured.jpg/png` to your page's folder then add a caption below.
+image:
+  caption: ''
+  focal_point: ''
+  preview_only: false
+
+# Associated Projects (optional).
+#   Associate this publication with one or more of your projects.
+#   Simply enter your project's folder or file name without extension.
+#   E.g. `projects: ['internal-project']` links to `content/project/internal-project/index.md`.
+#   Otherwise, set `projects: []`.
+projects: []
+links:
+- name: arXiv
+  url: https://arxiv.org/abs/2412.10095
+---

content/en/publication/biderman-2024-lmevaluation/cite.bib

@@ -0,0 +1,12 @@
+@article{biderman2024lmevaluation,
+ abstract = {Effective evaluation of language models remains an open challenge in NLP. Researchers and engineers face methodological issues such as the sensitivity of models to evaluation setup, difficulty of proper comparisons across methods, and the lack of reproducibility and transparency. In this paper we draw on three years of experience in evaluating large language models to provide guidance and lessons for researchers. First, we provide an overview of common challenges faced in language model evaluation. Second, we delineate best practices for addressing or lessening the impact of these challenges on research. Third, we present the Language Model Evaluation Harness (lm-eval): an open source library for independent, reproducible, and extensible evaluation of language models that seeks to address these issues. We describe the features of the library as well as case studies in which the library has been used to alleviate these methodological concerns.},
+ archiveprefix = {arXiv},
+ author = {Stella Biderman and Hailey Schoelkopf and Lintang Sutawika and Leo Gao and Jonathan Tow and Baber Abbasi and Alham Fikri Aji and Pawan Sasanka Ammanamanchi and Sidney Black and Jordan Clive and Anthony DiPofi and Julen Etxaniz and Benjamin Fattori and Jessica Zosa Forde and Charles Foster and Jeffrey Hsu and Mimansa Jaiswal and Wilson Y. Lee and Haonan Li and Charles Lovering and Niklas Muennighoff and Ellie Pavlick and Jason Phang and Aviya Skowron and Samson Tan and Xiangru Tang and Kevin A. Wang and Genta Indra Winata and François Yvon and Andy Zou},
+ date = {2024-05-23},
+ eprint = {2405.14782},
+ keywords = {Natural Language Processing, Large Language Models, Deep Learning, Evaluation, Reproducibility},
+ primaryclass = {cs.CL},
+ title = {Lessons from the Trenches on Reproducible Evaluation of Language Models},
+ url = {https://arxiv.org/abs/2405.14782},
+ year = {2024}
+}

content/en/publication/biderman-2024-lmevaluation/index.md

@@ -0,0 +1,117 @@
+---
+title: Lessons from the Trenches on Reproducible Evaluation of Language Models
+
+# Authors
+# A YAML list of author names
+# If you created a profile for a user (e.g. the default `admin` user at `content/authors/admin/`), 
+# write the username (folder name) here, and it will be replaced with their full name and linked to their profile.
+authors:
+- Stella Biderman
+- Hailey Schoelkopf
+- Lintang Sutawika
+- Leo Gao
+- Jonathan Tow
+- Baber Abbasi
+- Alham Fikri Aji
+- Pawan Sasanka Ammanamanchi
+- Sidney Black
+- Jordan Clive
+- Anthony DiPofi
+- Julen Etxaniz
+- Benjamin Fattori
+- Jessica Zosa Forde
+- Charles Foster
+- Jeffrey Hsu
+- Mimansa Jaiswal
+- Wilson Y. Lee
+- Haonan Li
+- Charles Lovering
+- Niklas Muennighoff
+- Ellie Pavlick
+- Jason Phang
+- Aviya Skowron
+- Samson Tan
+- Xiangru Tang
+- Kevin A. Wang
+- Genta Indra Winata
+- François Yvon
+- Andy Zou
+
+# Author notes (such as 'Equal Contribution')
+# A YAML list of notes for each author in the above `authors` list
+author_notes: []
+
+date: '2024-05-23'
+
+# Date to publish webpage (NOT necessarily Bibtex publication's date).
+publishDate: '2025-01-03T18:58:39.104447Z'
+
+# Publication type.
+# A single CSL publication type but formatted as a YAML list (for Hugo requirements).
+publication_types:
+- article-journal
+
+# Publication name and optional abbreviated publication name.
+publication: ''
+publication_short: ''
+
+doi: ''
+
+abstract: 'Effective evaluation of language models remains an open challenge in NLP.
+  Researchers and engineers face methodological issues such as the sensitivity of
+  models to evaluation setup, difficulty of proper comparisons across methods, and
+  the lack of reproducibility and transparency. In this paper we draw on three years
+  of experience in evaluating large language models to provide guidance and lessons
+  for researchers. First, we provide an overview of common challenges faced in language
+  model evaluation. Second, we delineate best practices for addressing or lessening
+  the impact of these challenges on research. Third, we present the Language Model
+  Evaluation Harness (lm-eval): an open source library for independent, reproducible,
+  and extensible evaluation of language models that seeks to address these issues.
+  We describe the features of the library as well as case studies in which the library
+  has been used to alleviate these methodological concerns.'
+
+# Summary. An optional shortened abstract.
+summary: ''
+
+tags:
+- Natural Language Processing
+- Large Language Models
+- Deep Learning
+- Evaluation
+- Reproducibility
+
+# Display this page in a list of Featured pages?
+featured: true
+
+# Links
+url_pdf: https://arxiv.org/pdf/2405.14782
+url_code: https://github.com/EleutherAI/lm-evaluation-harness
+url_dataset: ''
+url_poster: ''
+url_project: ''
+url_slides: ''
+url_source: ''
+url_video: ''
+
+# Custom links (uncomment lines below)
+# links:
+# - name: Custom Link
+#   url: http://example.org
+
+# Publication image
+# Add an image named `featured.jpg/png` to your page's folder then add a caption below.
+image:
+  caption: ''
+  focal_point: ''
+  preview_only: false
+
+# Associated Projects (optional).
+#   Associate this publication with one or more of your projects.
+#   Simply enter your project's folder or file name without extension.
+#   E.g. `projects: ['internal-project']` links to `content/project/internal-project/index.md`.
+#   Otherwise, set `projects: []`.
+projects: []
+links:
+- name: arXiv
+  url: https://arxiv.org/abs/2405.14782
+---

content/en/publication/etxaniz-2024-bertaqa/cite.bib

@@ -0,0 +1,13 @@
+@article{etxaniz2024bertaqa,
+ abstract = {Large Language Models (LLMs) exhibit extensive knowledge about the world, but most evaluations have been limited to global or anglocentric subjects. This raises the question of how well these models perform on topics relevant to other cultures, whose presence on the web is not that prominent. To address this gap, we introduce BertaQA, a multiple-choice trivia dataset that is parallel in English and Basque. The dataset consists of a local subset with questions pertinent to the Basque culture, and a global subset with questions of broader interest. We find that state-of-the-art LLMs struggle with local cultural knowledge, even as they excel on global topics. However, we show that continued pre-training in Basque significantly improves the models' performance on Basque culture, even when queried in English. To our knowledge, this is the first solid evidence of knowledge transfer from a low-resource to a high-resource language. Our analysis sheds light on the complex interplay between language and knowledge, and reveals that some prior findings do not fully hold when reassessed on local topics. Our dataset and evaluation code are available under open licenses at https://github.com/juletx/BertaQA.},
+ archiveprefix = {arXiv},
+ author = {Julen Etxaniz and Gorka Azkune and Aitor Soroa and Oier Lopez de Lacalle and Mikel Artetxe},
+ booktitle = {NeurIPS Dataset and Benchmarks 2024},
+ date = {2024-06-11},
+ eprint = {2406.07302},
+ keywords = {Natural Language Processing, Large Language Models, Deep Learning, Evaluation, Multilinguality, Culture, Basque},
+ primaryclass = {cs.CL},
+ title = {BertaQA: How Much Do Language Models Know About Local Culture?},
+ url = {https://arxiv.org/abs/2406.07302},
+ year = {2024}
+}

content/en/publication/etxaniz-2024-bertaqa/index.md

@@ -0,0 +1,97 @@
+---
+title: 'BertaQA: How Much Do Language Models Know About Local Culture?'
+
+# Authors
+# A YAML list of author names
+# If you created a profile for a user (e.g. the default `admin` user at `content/authors/admin/`), 
+# write the username (folder name) here, and it will be replaced with their full name and linked to their profile.
+authors:
+- Julen Etxaniz
+- Gorka Azkune
+- Aitor Soroa
+- Oier Lopez de Lacalle
+- Mikel Artetxe
+
+# Author notes (such as 'Equal Contribution')
+# A YAML list of notes for each author in the above `authors` list
+author_notes: []
+
+date: '2024-06-11'
+
+# Date to publish webpage (NOT necessarily Bibtex publication's date).
+publishDate: '2025-01-03T18:57:04.363466Z'
+
+# Publication type.
+# A single CSL publication type but formatted as a YAML list (for Hugo requirements).
+publication_types:
+- article-journal
+
+# Publication name and optional abbreviated publication name.
+publication: '*NeurIPS Datasets and Benchmarks 2024*'
+publication_short: ''
+
+doi: ''
+
+abstract: Large Language Models (LLMs) exhibit extensive knowledge about the world,
+  but most evaluations have been limited to global or anglocentric subjects. This
+  raises the question of how well these models perform on topics relevant to other
+  cultures, whose presence on the web is not that prominent. To address this gap,
+  we introduce BertaQA, a multiple-choice trivia dataset that is parallel in English
+  and Basque. The dataset consists of a local subset with questions pertinent to the
+  Basque culture, and a global subset with questions of broader interest. We find
+  that state-of-the-art LLMs struggle with local cultural knowledge, even as they
+  excel on global topics. However, we show that continued pre-training in Basque significantly
+  improves the models' performance on Basque culture, even when queried in English.
+  To our knowledge, this is the first solid evidence of knowledge transfer from a
+  low-resource to a high-resource language. Our analysis sheds light on the complex
+  interplay between language and knowledge, and reveals that some prior findings do
+  not fully hold when reassessed on local topics. Our dataset and evaluation code
+  are available under open licenses at https://github.com/juletx/BertaQA.
+
+# Summary. An optional shortened abstract.
+summary: ''
+
+tags:
+- Natural Language Processing
+- Large Language Models
+- Deep Learning
+- Evaluation
+- Multilinguality
+- Culture
+- Basque
+
+# Display this page in a list of Featured pages?
+featured: true
+
+# Links
+url_pdf: https://arxiv.org/pdf/2406.07302
+url_code: https://github.com/juletx/BertaQA
+url_dataset: https://huggingface.co/datasets/HiTZ/BertaQA
+url_poster: ''
+url_project: ''
+url_slides: ''
+url_source: ''
+url_video: ''
+
+# Custom links (uncomment lines below)
+# links:
+# - name: Custom Link
+#   url: http://example.org
+
+# Publication image
+# Add an image named `featured.jpg/png` to your page's folder then add a caption below.
+image:
+  caption: ''
+  focal_point: ''
+  preview_only: false
+
+# Associated Projects (optional).
+#   Associate this publication with one or more of your projects.
+#   Simply enter your project's folder or file name without extension.
+#   E.g. `projects: ['internal-project']` links to `content/project/internal-project/index.md`.
+#   Otherwise, set `projects: []`.
+projects: []
+links:
+- name: arXiv
+  url: https://arxiv.org/abs/2406.07302
+---

content/en/publication/pensa-2024-gita-4-calamita/cite.bib

@@ -0,0 +1,16 @@
+@article{pensa2024gita4calamita,
+ abstract = {In the context of the CALAMITA Challenge, we investigate the physical commonsense reasoning capabilities of large language models (LLMs) and introduce a methodology to assess their understanding of the physical world. To this end, we use a test set designed to evaluate physical commonsense reasoning in LLMs for the Italian language. We present a tiered dataset,
+named the Graded Italian Annotated dataset (GITA), which is written and annotated by a professional linguist. This dataset
+enables us to focus on three distinct levels of commonsense understanding. Our benchmark aims to evaluate three specific
+tasks: identifying plausible and implausible stories within our dataset, identifying the conflict that generates an implausible
+story, and identifying the physical states that make a story implausible. We perform these tasks using LLAMA3, Gemma2
+and Mistral. Our findings reveal that, although the models may excel at high-level classification tasks, their reasoning is
+inconsistent and unverifiable, as they fail to capture intermediate evidence.},
+ author = {Pensa, Giulia and Azurmendi, Ekhi and Etxaniz, Julen and Altuna, Begoña and Gonzalez-Dios, Itziar},
+ booktitle = {CLiC-it 2024},
+ date = {2024-12-06},
+ keywords = {Natural Language Processing, Large Language Models, Deep Learning, Evaluation, Commonsense Reasoning, Italian},
+ title = {GITA4CALAMITA - Evaluating the Physical Commonsense Understanding of Italian LLMs in a Multi-layered Approach: A CALAMITA Challenge},
+ url = {https://ceur-ws.org/Vol-3878/127_calamita_long.pdf},
+ year = {2024}
+}