Why don't you do it right? analysing annotators' disagreement in subjective tasks

M Sandri, E Leonardelli, S Tonelli… - Proceedings of the 17th …, 2023‏ - aclanthology.org
Annotators' disagreement in linguistic data has been recently the focus of multiple initiatives
aimed at raising awareness on issues related to 'majority voting'when aggregating diverging …

Deep dominance-how to properly compare deep neural models

R Dror, S Shlomov, R Reichart - … of the 57th Annual Meeting of the …, 2019‏ - aclanthology.org
Abstract Comparing between Deep Neural Network (DNN) models based on their
performance on unseen data is crucial for the progress of the NLP field. However, these …

A noisy elephant in the room: Is your out-of-distribution detector robust to label noise?

G Humblot-Renaux, S Escalera… - Proceedings of the …, 2024‏ - openaccess.thecvf.com
The ability to detect unfamiliar or unexpected images is essential for safe deployment of
computer vision systems. In the context of classification the task of detecting images outside …

Calibrating large language models using their generations only

D Ulmer, M Gubri, H Lee, S Yun, SJ Oh - ar** llm-based task-oriented dialogue agents via self-talk
D Ulmer, E Mansimov, K Lin, J Sun, X Gao… - arxiv preprint arxiv …, 2024‏ - arxiv.org
Large language models (LLMs) are powerful dialogue agents, but specializing them towards
fulfilling a specific function can be challenging. Instructing tuning, ie tuning models on …

Context-aware attention layers coupled with optimal transport domain adaptation and multimodal fusion methods for recognizing dementia from spontaneous speech

L Ilias, D Askounis - Knowledge-Based Systems, 2023‏ - Elsevier
Alzheimer's disease (AD) constitutes a complex neurocognitive disease and is the main
cause of dementia. Although many studies have been proposed targeting at diagnosing …

[HTML][HTML] Are you sure it's an artifact? Artifact detection and uncertainty quantification in histological images

N Kanwal, M López-Pérez, U Kiraz… - … Medical Imaging and …, 2024‏ - Elsevier
Modern cancer diagnostics involves extracting tissue specimens from suspicious areas and
conducting histotechnical procedures to prepare a digitized glass slide, called Whole Slide …

Exploring predictive uncertainty and calibration in NLP: A study on the impact of method & data scarcity

D Ulmer, J Frellsen, C Hardmeier - arxiv preprint arxiv:2210.15452, 2022‏ - arxiv.org
We investigate the problem of determining the predictive confidence (or, conversely,
uncertainty) of a neural classifier through the lens of low-resource languages. By training …

Not all layers are equally as important: Every layer counts BERT

LGG Charpentier, D Samuel - arxiv preprint arxiv:2311.02265, 2023‏ - arxiv.org
This paper introduces a novel modification of the transformer architecture, tailored for the
data-efficient pretraining of language models. This aspect is evaluated by participating in the …

[HTML][HTML] A multivariable sensor-agnostic framework for spatio-temporal air quality forecasting based on Deep Learning

II Prado-Rujas, A García-Dopico, E Serrano… - … Applications of Artificial …, 2024‏ - Elsevier
Recently, air quality has become a major concern for the protection of the environment and
the well-being of people. Air pollution is a key proxy of the quality of life in any city and is …