arXiv subjects

Brian Davis

Publications and source records attributed to Brian Davis.

At least 19 recordsLinked to original sources

arXiv2026

Validating DBpedia Triple Sets for Natural Language Generation

We present a study of the quality of individual DBpedia triples from the perspective of Natural Language Generation, and propose and evaluate an approach for collecting entity-specific triple sets that filters out questionable triples while minimizing the loss of correct ones. We show in an evaluation against manually annotated data that with validation rules, it is possible to reach 98% precision in triple selection, and with improvements to a few Property definitions, it is possible to improve recall by 40% without harming precision.

cs.CL↗

arXiv2026

MoirfEolas and Cr\'iochScore: Developing Resources for and the Evaluation of Tokenization Alignment with Irish Morphology

This paper presents new tokenization resources for Irish and evaluation measures of alignment with the morphological boundaries of the language. We present MoirfEolas, a dataset of over 35,000 Irish words mapped to their respective eclipses, prefixes and suffixes as well as an evaluation metric Cr\'iochScore, that evaluates the alignment of tokenizations with the morphological boundaries present in MoirfEolas. We evaluate common tokenization algorithms using Cr\'iochScore as well as intrinsic metrics present in the tokenization literature. We find that the Unigram Language Model aligns with Irish morphology more often than the other algorithms evaluated. We also find trade-offs between morphological-alignment of tokenization with both compression as well as vocabulary efficiency, providing practical insights for Irish natural language processing development. This dataset contributes towards combating the Irish language's low-resource status; moreover, the construction process reported in this paper can be emulated by other languages to create specialised morphological resources.

cs.CL↗

arXiv2025

SynBullying: A Multi LLM Synthetic Conversational Dataset for Cyberbullying Detection

We introduce SynBullying, a synthetic multi-LLM conversational dataset for studying and detecting cyberbullying (CB). SynBullying provides a scalable and ethically safe alternative to human data collection by leveraging large language models (LLMs) to simulate realistic bullying interactions. The dataset offers (i) conversational structure, capturing multi-turn exchanges rather than isolated posts; (ii) context-aware annotations, where harmfulness is assessed within the conversational flow considering context, intent, and discourse dynamics; and (iii) fine-grained labeling, covering various CB categories for detailed linguistic and behavioral analysis. We evaluate SynBullying across five dimensions, including conversational structure, lexical patterns, sentiment/toxicity, role dynamics, harm intensity, and CB-type distribution. We further examine its utility by testing its performance as standalone training data and as an augmentation source for CB classification.

cs.AI↗

arXiv2025

Long-context Reference-based MT Quality Estimation

In this paper, we present our submission to the Tenth Conference on Machine Translation (WMT25) Shared Task on Automated Translation Quality Evaluation. Our systems are built upon the COMET framework and trained to predict segment-level Error Span Annotation (ESA) scores using augmented long-context data. To construct long-context training data, we concatenate in-domain, human-annotated sentences and compute a weighted average of their scores. We integrate multiple human judgment datasets (MQM, SQM, and DA) by normalising their scales and train multilingual regression models to predict quality scores from the source, hypothesis, and reference translations. Experimental results show that incorporating long-context information improves correlations with human judgments compared to models trained only on short segments.

cs.CL↗

arXiv2025

Synthetic vs. Gold: The Role of LLM Generated Labels and Data in Cyberbullying Detection

Cyberbullying (CB) presents a pressing threat, especially to children, underscoring the urgent need for robust detection systems to ensure online safety. While large-scale datasets on online abuse exist, there remains a significant gap in labeled data that specifically reflects the language and communication styles used by children. The acquisition of such data from vulnerable populations, such as children, is challenging due to ethical, legal and technical barriers. Moreover, the creation of these datasets relies heavily on human annotation, which not only strains resources but also raises significant concerns due to annotators exposure to harmful content. In this paper, we address these challenges by leveraging Large Language Models (LLMs) to generate synthetic data and labels. Our experiments demonstrate that synthetic data enables BERT-based CB classifiers to achieve performance close to that of those trained on fully authentic datasets (75.8% vs. 81.5% accuracy). Additionally, LLMs can effectively label authentic yet unlabeled data, allowing BERT classifiers to attain a comparable performance level (79.1% vs. 81.5% accuracy). These results highlight the potential of LLMs as a scalable, ethical, and cost-effective solution for generating data for CB detection.

cs.CL↗

arXiv2025

Annotating Compositionality Scores for Irish Noun Compounds is Hard Work

Noun compounds constitute a challenging construction for NLP applications, given their variability in idiomaticity and interpretation. In this paper, we present an analysis of compound nouns identified in Irish text of varied domains by expert annotators, focusing on compositionality as a key feature, but also domain specificity, as well as familiarity and confidence of the annotator giving the ratings. Our findings and the discussion that ensued contributes towards a greater understanding of how these constructions appear in Irish language, and how they might be treated separately from English noun compounds.

cs.CL↗

arXiv2024

A Systematic Review of Data-to-Text NLG

This systematic review undertakes a comprehensive analysis of current research on data-to-text generation, identifying gaps, challenges, and future directions within the field. Relevant literature in this field on datasets, evaluation metrics, application areas, multilingualism, language models, and hallucination mitigation methods is reviewed. Various methods for producing high-quality text are explored, addressing the challenge of hallucinations in data-to-text generation. These methods include re-ranking, traditional and neural pipeline architecture, planning architectures, data cleaning, controlled generation, and modification of models and training techniques. Their effectiveness and limitations are assessed, highlighting the need for universally applicable strategies to mitigate hallucinations. The review also examines the usage, popularity, and impact of datasets, alongside evaluation metrics, with an emphasis on both automatic and human assessment. Additionally, the evolution of data-to-text models, particularly the widespread adoption of transformer models, is discussed. Despite advancements in text quality, the review emphasizes the importance of research in low-resourced languages and the engineering of datasets in these languages to promote inclusivity. Finally, several application domains of data-to-text are highlighted, emphasizing their relevance in such domains. Overall, this review serves as a guiding framework for fostering innovation and advancing data-to-text generation.

cs.CL↗

arXiv2023

DeepQC: A Deep Learning System for Automatic Quality Control of In-situ Soil Moisture Sensor Time Series Data

Amidst changing climate, real-time soil moisture monitoring is vital for the development of in-season decision support tools to help farmers manage weather related risks. Precision Sustainable Agriculture (PSA) recently established a real-time soil moisture monitoring network across the central, Midwest, and eastern U.S., but field-scale sensor observations often come with data gaps and anomalies. To maintain the data quality needed for development of decision tools, a quality control system is necessary. The International Soil Moisture Network (ISMN) introduced the Flagit module for anomaly detection in soil moisture observations. However, under certain conditions, Flagit's quality control approaches may underperform in identifying anomalies. Recently deep learning methods have been successfully applied to detect anomalies in time series data in various disciplines. However, their use in agriculture has not been yet investigated. This study focuses on developing a Bi-directional Long Short-Term Memory (LSTM) model, referred to as DeepQC, to identify anomalies in soil moisture data. Manual flagged PSA observations were used for training, validation, and testing the model, following an 80:10:10 split. The study then compared the DeepQC and Flagit based estimates to assess their relative performance. Flagit corrected flagged 95.5% of the corrected observations and 50.3% of the anomaly observations, indicating its limitations in identifying anomalies. On the other hand, the DeepQC correctly flagged 99.7% of the correct observations and 95.6% of the anomalies in significantly less time, demonstrating its superiority over Flagit approach. Importantly, DeepQC's performance remained consistent regardless of the number of anomalies. Given the promising results obtained with the DeepQC, future studies will focus on implementing this model on national and global soil moisture networks.

cs.LG↗

arXiv2022

End-to-end Document Recognition and Understanding with Dessurt

We introduce Dessurt, a relatively simple document understanding transformer capable of being fine-tuned on a greater variety of document tasks than prior methods. It receives a document image and task string as input and generates arbitrary text autoregressively as output. Because Dessurt is an end-to-end architecture that performs text recognition in addition to the document understanding, it does not require an external recognition model as prior methods do. Dessurt is a more flexible model than prior methods and is able to handle a variety of document domains and tasks. We show that this model is effective at 9 different dataset-task combinations.

cs.CV↗

arXiv2021

Optimal Inverted Landing in a Small Aerial Robot with Varied Approach Velocities and Landing Gear Designs

Inverted landing is a challenging feat to perform in aerial robots, especially without external positioning. However, it is routinely performed by biological fliers such as bees, flies, and bats. Our previous observations of landing behaviors in flies suggest an open-loop causal relationship between their putative visual cues and the kinematics of the aerial maneuvers executed. For example, the degree of rotational maneuver (the amount of body inversion prior to touchdown) and the amount of leg-assisted body swing both depend on the flies' initial body states while approaching the ceiling. In this work, inspired by the inverted landing behavior of flies, we used a physics-based simulation with experimental validation to systematically investigate how optimized inverted landing maneuvers depend on the initial approach velocities with varied magnitude and direction. This was done by analyzing the putative visual cues (that can be derived from onboard measurements) during optimal maneuvering trajectories. We identified a three-dimensional policy region, from which a mapping to a global inverted landing policy can be developed without the use of external positioning data. Through simulation, we also investigated the effects of an array of landing gear designs on the optimized landing performance and identified their advantages and disadvantages. The above results have been partially validated using limited experimental testing and will continue to inform and guide our future experiments, for example by applying the calculated global policy.

cs.RO↗

arXiv2021

Visual FUDGE: Form Understanding via Dynamic Graph Editing

We address the problem of form understanding: finding text entities and the relationships/links between them in form images. The proposed FUDGE model formulates this problem on a graph of text elements (the vertices) and uses a Graph Convolutional Network to predict changes to the graph. The initial vertices are detected text lines and do not necessarily correspond to the final text entities, which can span multiple lines. Also, initial edges contain many false-positive relationships. FUDGE edits the graph structure by combining text segments (graph vertices) and pruning edges in an iterative fashion to obtain the final text entities and relationships. While recent work in this area has focused on leveraging large-scale pre-trained Language Models (LM), FUDGE achieves almost the same level of entity linking performance on the FUNSD dataset by learning only visual features from the (small) provided training set. FUDGE can be applied on forms where text recognition is difficult (e.g. degraded or historical forms) and on forms in resource-poor languages where pre-training such LMs is challenging. FUDGE is state-of-the-art on the historical NAF dataset.

cs.CV↗

arXiv2020

Over a Decade of Social Opinion Mining: A Systematic Review

Social media popularity and importance is on the increase due to people using it for various types of social interaction across multiple channels. This systematic review focuses on the evolving research area of Social Opinion Mining, tasked with the identification of multiple opinion dimensions, such as subjectivity, sentiment polarity, emotion, affect, sarcasm and irony, from user-generated content represented across multiple social media platforms and in various media formats, like text, image, video and audio. Through Social Opinion Mining, natural language can be understood in terms of the different opinion dimensions, as expressed by humans. This contributes towards the evolution of Artificial Intelligence which in turn helps the advancement of several real-world use cases, such as customer service and decision making. A thorough systematic review was carried out on Social Opinion Mining research which totals 485 published studies and spans a period of twelve years between 2007 and 2018. The in-depth analysis focuses on the social media platforms, techniques, social datasets, language, modality, tools and technologies, and other aspects derived. Social Opinion Mining can be utilised in many application areas, ranging from marketing, advertising and sales for product/service management, and in multiple domains and industries, such as politics, technology, finance, healthcare, sports and government. The latest developments in Social Opinion Mining beyond 2018 are also presented together with future research directions, with the aim of leaving a wider academic and societal impact in several real-world applications.

cs.CL↗

arXiv2020

Text and Style Conditioned GAN for Generation of Offline Handwriting Lines

This paper presents a GAN for generating images of handwritten lines conditioned on arbitrary text and latent style vectors. Unlike prior work, which produce stroke points or single-word images, this model generates entire lines of offline handwriting. The model produces variable-sized images by using style vectors to determine character widths. A generator network is trained with GAN and autoencoder techniques to learn style, and uses a pre-trained handwriting recognition network to induce legibility. A study using human evaluators demonstrates that the model produces images that appear to be written by a human. After training, the encoder network can extract a style vector from an image, allowing images in a similar style to be generated, but with arbitrary text.

cs.CV↗

arXiv2020

Split Optimization for Protein/Ligand Binding Models

In this paper, we investigate potential biases in datasets used to make drug binding predictions using machine learning. We investigate a recently published metric called the Asymmetric Validation Embedding (AVE) bias which is used to quantify this bias and detect overfitting. We compare it to a slightly revised version and introduce a new weighted metric. We find that the new metrics allow to quantify overfitting while not overly limiting training data and produce models with greater predictive value.

q-bio.BM↗

arXiv2019

Deep Visual Template-Free Form Parsing

Automatic, template-free extraction of information from form images is challenging due to the variety of form layouts. This is even more challenging for historical forms due to noise and degradation. A crucial part of the extraction process is associating input text with pre-printed labels. We present a learned, template-free solution to detecting pre-printed text and input text/handwriting and predicting pair-wise relationships between them. While previous approaches to this problem have been focused on clean images and clear layouts, we show our approach is effective in the domain of noisy, degraded, and varied form images. We introduce a new dataset of historical form images (late 1800s, early 1900s) for training and validating our approach. Our method uses a convolutional network to detect pre-printed text and input text lines. We pool features from the detection network to classify possible relationships in a language-agnostic way. We show that our proposed pairing method outperforms heuristic rules and that visual features are critical to obtaining high accuracy.

cs.CV↗

arXiv2019

On Network Science and Mutual Information for Explaining Deep Neural Networks

In this paper, we present a new approach to interpret deep learning models. By coupling mutual information with network science, we explore how information flows through feedforward networks. We show that efficiently approximating mutual information allows us to create an information measure that quantifies how much information flows between any two neurons of a deep learning model. To that end, we propose NIF, Neural Information Flow, a technique for codifying information flow that exposes deep learning model internals and provides feature attributions.

cs.LG↗

arXiv2019

Antichain Simplices

To each lattice simplex $\Delta$ we associate a poset encoding the additive structure of lattice points in the fundamental parallelepiped for $\Delta$. When this poset is an antichain, we say $\Delta$ is antichain. To each partition $\lambda$ of $n$, we associate a lattice simplex $\Delta_\lambda$ having one unimodular facet, and we investigate their associated posets. We give a number-theoretic characterization of the relations in these posets, as well as a simplified characterization in the case where each part of $\lambda$ is relatively prime to $n-1$. We use these characterizations to experimentally study $\Delta_\lambda$ for all partitions of $n$ with $n\leq 73$. We also investigate the structure of these posets when $\lambda$ has only one or two distinct parts. Finally, we explain how this work relates to Poincar\'e series for the semigroup algebra associated to $\Delta$, and we prove that this series is rational when $\Delta$ is antichain.

math.CO↗

arXiv2018

Language Model Supervision for Handwriting Recognition Model Adaptation

Training state-of-the-art offline handwriting recognition (HWR) models requires large labeled datasets, but unfortunately such datasets are not available in all languages and domains due to the high cost of manual labeling.We address this problem by showing how high resource languages can be leveraged to help train models for low resource languages.We propose a transfer learning methodology where we adapt HWR models trained on a source language to a target language that uses the same writing script.This methodology only requires labeled data in the source language, unlabeled data in the target language, and a language model of the target language. The language model is used in a bootstrapping fashion to refine predictions in the target language for use as ground truth in training the model.Using this approach we demonstrate improved transferability among French, English, and Spanish languages using both historical and modern handwriting datasets. In the best case, transferring with the proposed methodology results in character error rates nearly as good as full supervised training.

cs.CV↗