arXiv subjects

Marian-Andrei Rizoiu

Publications and source records attributed to Marian-Andrei Rizoiu.

At least 37 records · Page 2Linked to original sources

arXiv2025

X-Troll: eXplainable Detection of State-Sponsored Information Operations Agents

State-sponsored trolls, malicious actors who deploy sophisticated linguistic manipulation in coordinated information campaigns, posing threats to online discourse integrity. While Large Language Models (LLMs) achieve strong performance on general natural language processing (NLP) tasks, they struggle with subtle propaganda detection and operate as ``black boxes'', providing no interpretable insights into manipulation strategies. This paper introduces X-Troll, a novel framework that bridges this gap by integrating explainable adapter-based LLMs with expert-derived linguistic knowledge to detect state-sponsored trolls and provide human-readable explanations for its decisions. X-Troll incorporates appraisal theory and propaganda analysis through specialized LoRA adapters, using dynamic gating to capture campaign-specific discourse patterns in coordinated information operations. Experiments on real-world data demonstrate that our linguistically-informed approach shows strong performance compared with both general LLM baselines and existing troll detection models in accuracy while providing enhanced transparency through expert-grounded explanations that reveal the specific linguistic strategies used by state-sponsored actors. X-Troll source code is available at: https://github.com/ltian678/xtroll_source/.

cs.CL↗

arXiv2025

Practical Guidelines for Ideology Detection Pipelines and Psychosocial Applications

Online ideology detection is crucial for downstream tasks, like countering ideologically motivated violent extremism and modeling opinion dynamics. However, two significant issues arise in practitioners' deployment. Firstly, gold-standard training data is prohibitively labor-intensive to collect and has limited reusability beyond its collection context (i.e., time, location, and platform). Secondly, to circumvent expense, researchers employ ideological signals (such as hashtags shared). Unfortunately, these signals' annotation requirements and context transferability are largely unknown, and the bias they induce remains unquantified. This study provides guidelines for practitioners requiring real-time detection of left, right, and extreme ideologies in large-scale online settings. We propose a framework for pipeline constructions, describing ideology signals by their associated labor and context transferability. We evaluate many constructions, quantifying the bias associated with signals and describing a pipeline that outperforms state-of-the-art methods ($0.95$ AUC ROC). We showcase the capabilities of our pipeline on five datasets containing more than 1.12 million users. We set out to investigate whether the findings in the psychosocial literature, developed for the offline environment, apply to the online setting. We evaluate at scale several psychosocial hypotheses that delineate ideologies concerning morality, grievance, nationalism, and dichotomous thinking. We find that right-wing ideologies use more vice-moral language, have more grievance-filled language, exhibit increased black-and-white thinking patterns, and have a greater association with national flags. This research empowers practitioners with guidelines for ideology detection, and case studies for its application, fostering a safer and better understood digital landscape.

cs.CY↗

arXiv2025

Mario at EXIST 2025: A Simple Gateway to Effective Multilingual Sexism Detection

This paper presents our approach to EXIST 2025 Task 1, addressing text-based sexism detection in English and Spanish tweets through hierarchical Low-Rank Adaptation (LoRA) of Llama 3.1 8B. Our method introduces conditional adapter routing that explicitly models label dependencies across three hierarchically structured subtasks: binary sexism identification, source intention detection, and multilabel sexism categorization. Unlike conventional LoRA applications that target only attention layers, we apply adaptation to all linear transformations, enhancing the model's capacity to capture task-specific patterns. In contrast to complex data processing and ensemble approaches, we show that straightforward parameter-efficient fine-tuning achieves strong performance. We train separate LoRA adapters (rank=16, QLoRA 4-bit) for each subtask using unified multilingual training that leverages Llama 3.1's native bilingual capabilities. The method requires minimal preprocessing and uses standard supervised learning. Our multilingual training strategy eliminates the need for separate language-specific models, achieving 1.7-2.4\% F1 improvements through cross-lingual transfer. With only 1.67\% trainable parameters compared to full fine-tuning, our approach reduces training time by 75\% and model storage by 98\%, while achieving competitive performance across all subtasks (ICM-Hard: 0.6774 for binary classification, 0.4991 for intention detection, 0.6519 for multilabel categorization).

cs.CL↗

arXiv2025

Skill-Driven Certification Pathways: Measuring Industry Training Impact on Graduate Employability

Australia faces a critical technology skills shortage, requiring approximately $52,000$ new technology professionals annually by 2030, while confronting a widening gap between employer requirements and graduate capabilities. With only $1\%$ of technology graduates considered immediately work-ready, traditional educational pathways alone prove insufficient to meet industry demands. This research examines how industry certifications, such as Microsoft's AI-900 (Azure AI Fundamentals), can bridge this critical skills gap. We propose a novel, data-driven methodology that quantitatively measures skill alignment between educational offerings and job market requirements by analysing over 2.5 million job advertisements from Australia, the US, and the UK, mapping extracted skills to industry taxonomies using the Vectorised Skills Space Method. Our findings reveal that combining university degrees with targeted industry certifications significantly enhances employability for technology roles. The Bachelor of Computer Science with AI major combined with AI-900 certification achieved the highest absolute skill similarity score for Machine Learning Engineer positions. Surprisingly, the largest improvements when augmented with AI certifications are experiences by non-technical degrees--such as nursing nursing--with up to $9,296\%$ percentage improvements in alignment with Machine Learning Engineer roles. Our results challenge conventional assumptions about technology career pathways. They can provide actionable insights for educational institutions seeking evidence-based curriculum design, students requiring strategic certification guidance, and employers recognising potential in candidates from non-traditional backgrounds who have obtained relevant certifications.

cs.CY↗

arXiv2025

Estimating Online Influence Needs Causal Modeling! Counterfactual Analysis of Social Media Engagement

Understanding true influence in social media requires distinguishing correlation from causation--particularly when analyzing misinformation spread. While existing approaches focus on exposure metrics and network structures, they often fail to capture the causal mechanisms by which external temporal signals trigger engagement. We introduce a novel joint treatment-outcome framework that leverages existing sequential models to simultaneously adapt to both policy timing and engagement effects. Our approach adapts causal inference techniques from healthcare to estimate Average Treatment Effects (ATE) within the sequential nature of social media interactions, tackling challenges from external confounding signals. Through our experiments on real-world misinformation and disinformation datasets, we show that our models outperform existing benchmarks by 15--22% in predicting engagement across diverse counterfactual scenarios, including exposure adjustment, timing shifts, and varied intervention durations. Case studies on 492 social media users show our causal effect measure aligns strongly with the gold standard in influence estimation, the expert-based empirical influence.

cs.CL↗

arXiv2025

Signals from the Floods: AI-Driven Disaster Analysis through Multi-Source Data Fusion

Massive and diverse web data are increasingly vital for government disaster response, as demonstrated by the 2022 floods in New South Wales (NSW), Australia. This study examines how X (formerly Twitter) and public inquiry submissions provide insights into public behaviour during crises. We analyse more than 55,000 flood-related tweets and 1,450 submissions to identify behavioural patterns during extreme weather events. While social media posts are short and fragmented, inquiry submissions are detailed, multi-page documents offering structured insights. Our methodology integrates Latent Dirichlet Allocation (LDA) for topic modelling with Large Language Models (LLMs) to enhance semantic understanding. LDA reveals distinct opinions and geographic patterns, while LLMs improve filtering by identifying flood-relevant tweets using public submissions as a reference. This Relevance Index method reduces noise and prioritizes actionable content, improving situational awareness for emergency responders. By combining these complementary data streams, our approach introduces a novel AI-driven method to refine crisis-related social media content, improve real-time disaster response, and inform long-term resilience planning.

cs.CL↗

arXiv2025

Generalizing Hate Speech Detection Using Multi-Task Learning: A Case Study of Political Public Figures

Automatic identification of hateful and abusive content is vital in combating the spread of harmful online content and its damaging effects. Most existing works evaluate models by examining the generalization error on train-test splits on hate speech datasets. These datasets often differ in their definitions and labeling criteria, leading to poor generalization performance when predicting across new domains and datasets. This work proposes a new Multi-task Learning (MTL) pipeline that trains simultaneously across multiple hate speech datasets to construct a more encompassing classification model. Using a dataset-level leave-one-out evaluation (designating a dataset for testing and jointly training on all others), we trial the MTL detection on new, previously unseen datasets. Our results consistently outperform a large sample of existing work. We show strong results when examining the generalization error in train-test splits and substantial improvements when predicting on previously unseen datasets. Furthermore, we assemble a novel dataset, dubbed PubFigs, focusing on the problematic speech of American Public Political Figures. We crowdsource-label using Amazon MTurk more than $20,000$ tweets and machine-label problematic speech in all the $305,235$ tweets in PubFigs. We find that the abusive and hate tweeting mainly originates from right-leaning figures and relates to six topics, including Islam, women, ethnicity, and immigrants. We show that MTL builds embeddings that can simultaneously separate abusive from hate speech, and identify its topics.

cs.CL↗

arXiv2025

From Occupations to Tasks: A New Perspective on Automatability Prediction Using BERT

As automation technologies continue to advance at an unprecedented rate, concerns about job displacement and the future of work have become increasingly prevalent. While existing research has primarily focused on the potential impact of automation at the occupation level, there has been a lack of investigation into the automatability of individual tasks. This paper addresses this gap by proposing a BERT-based classifier to predict the automatability of tasks in the forthcoming decade at a granular level leveraging the context and semantics information of tasks. We leverage three public datasets: O*NET Task Statements, ESCO Skills, and Australian Labour Market Insights Tasks, and perform expert annotation. Our BERT-based classifier, fine-tuned on our task statement data, demonstrates superior performance over traditional machine learning models, neural network architectures, and other transformer models. Our findings also indicate that approximately 25.1% of occupations within the O*NET database are at substantial risk of automation, with a diverse spectrum of automation vulnerability across sectors. This research provides a robust tool for assessing the future impact of automation on the labor market, offering valuable insights for policymakers, workers, and industry leaders in the face of rapid technological advancement.

cs.CY↗

arXiv2025

Before It's Too Late: A State Space Model for the Early Prediction of Misinformation and Disinformation Engagement

In today's digital age, conspiracies and information campaigns can emerge rapidly and erode social and democratic cohesion. While recent deep learning approaches have made progress in modeling engagement through language and propagation models, they struggle with irregularly sampled data and early trajectory assessment. We present IC-Mamba, a novel state space model that forecasts social media engagement by modeling interval-censored data with integrated temporal embeddings. Our model excels at predicting engagement patterns within the crucial first 15-30 minutes of posting (RMSE 0.118-0.143), enabling rapid assessment of content reach. By incorporating interval-censored modeling into the state space framework, IC-Mamba captures fine-grained temporal dynamics of engagement growth, achieving a 4.72% improvement over state-of-the-art across multiple engagement metrics (likes, shares, comments, and emojis). Our experiments demonstrate IC-Mamba's effectiveness in forecasting both post-level dynamics and broader narrative patterns (F1 0.508-0.751 for narrative-level predictions). The model maintains strong predictive performance across extended time horizons, successfully forecasting opinion-level engagement up to 28 days ahead using observation windows of 3-10 days. These capabilities enable earlier identification of potentially problematic content, providing crucial lead time for designing and implementing countermeasures. Code is available at: https://github.com/ltian678/ic-mamba. An interactive dashboard demonstrating our results is available at: https://ic-mamba.behavioral-ds.science.

cs.CL↗

arXiv2025

Behavioral Homophily in Social Media via Inverse Reinforcement Learning: A Reddit Case Study

Online communities play a critical role in shaping societal discourse and influencing collective behavior in the real world. The tendency for people to connect with others who share similar characteristics and views, known as homophily, plays a key role in the formation of echo chambers which further amplify polarization and division. Existing works examining homophily in online communities traditionally infer it using content- or adjacency-based approaches, such as constructing explicit interaction networks or performing topic analysis. These methods fall short for platforms where interaction networks cannot be easily constructed and fail to capture the complex nature of user interactions across the platform. This work introduces a novel approach for quantifying user homophily. We first use an Inverse Reinforcement Learning (IRL) framework to infer users' policies, then use these policies as a measure of behavioral homophily. We apply our method to Reddit, conducting a case study across 5.9 million interactions over six years, demonstrating how this approach uncovers distinct behavioral patterns and user roles that vary across different communities. We further validate our behavioral homophily measure against traditional content-based homophily, offering a powerful method for analyzing social media dynamics and their broader societal implications. We find, among others, that users can behave very similarly (high behavioral homophily) when discussing entirely different topics like soccer vs e-sports (low topical homophily), and that there is an entire class of users on Reddit whose purpose seems to be to disagree with others.

cs.SI↗

arXiv2024

Linking Across Data Granularity: Fitting Multivariate Hawkes Processes to Partially Interval-Censored Data

The multivariate Hawkes process (MHP) is widely used for analyzing data streams that interact with each other, where events generate new events within their own dimension (via self-excitation) or across different dimensions (via cross-excitation). However, in certain applications, the timestamps of individual events in some dimensions are unobservable, and only event counts within intervals are known, referred to as partially interval-censored data. The MHP is unsuitable for handling such data since its estimation requires event timestamps. In this study, we introduce the Partially Censored Multivariate Hawkes Process (PCMHP), a novel point process which shares parameter equivalence with the MHP and can effectively model both timestamped and interval-censored data. We demonstrate the capabilities of the PCMHP using synthetic and real-world datasets. Firstly, we illustrate that the PCMHP can approximate MHP parameters and recover the spectral radius using synthetic event histories. Next, we assess the performance of the PCMHP in predicting YouTube popularity and find that the PCMHP outperforms the popularity estimation algorithm Hawkes Intensity Process (HIP). Comparing with the fully interval-censored HIP, we show that the PCMHP improves prediction performance by accounting for point process dimensions, particularly when there exist significant cross-dimension interactions. Lastly, we leverage the PCMHP to gain qualitative insights from a dataset comprising daily COVID-19 case counts from multiple countries and COVID-19-related news articles. By clustering the PCMHP-modeled countries, we unveil hidden interaction patterns between occurrences of COVID-19 cases and news reporting.

cs.LG↗

arXiv2024

What Drives Online Popularity: Author, Content or Sharers? Estimating Spread Dynamics with Bayesian Mixture Hawkes

The spread of content on social media is shaped by intertwining factors on three levels: the source, the content itself, and the pathways of content spread. At the lowest level, the popularity of the sharing user determines its eventual reach. However, higher-level factors such as the nature of the online item and the credibility of its source also play crucial roles in determining how widely and rapidly the online item spreads. In this work, we propose the Bayesian Mixture Hawkes (BMH) model to jointly learn the influence of source, content and spread. We formulate the BMH model as a hierarchical mixture model of separable Hawkes processes, accommodating different classes of Hawkes dynamics and the influence of feature sets on these classes. We test the BMH model on two learning tasks, cold-start popularity prediction and temporal profile generalization performance, applying to two real-world retweet cascade datasets referencing articles from controversial and traditional media publishers. The BMH model outperforms the state-of-the-art models and predictive baselines on both datasets and utilizes cascade- and item-level information better than the alternatives. Lastly, we perform a counter-factual analysis where we apply the trained publisher-level BMH models to a set of article headlines and show that effectiveness of headline writing style (neutral, clickbait, inflammatory) varies across publishers. The BMH model unveils differences in style effectiveness between controversial and reputable publishers, where we find clickbait to be notably more effective for reputable publishers as opposed to controversial ones, which links to the latter's overuse of clickbait.

cs.LG↗

arXiv2024

Misinformation is not about Bad Facts: An Analysis of the Production and Consumption of Fringe Content

What if misinformation is not an information problem at all? To understand the role of news publishers in potentially unintentionally propagating misinformation, we examine how far-right and fringe online groups share and leverage established legacy news media articles to advance their narratives. Our findings suggest that online fringe ideologies spread through the use of content that is consensus-based and "factually correct". We found that Australian news publishers with both moderate and far-right political leanings contain comparable levels of information completeness and quality; and furthermore, that far-right Twitter users often share from moderate sources. However, a stark difference emerges when we consider two additional factors: 1) the narrow topic selection of articles by far-right users, suggesting that they cherry pick only news articles that engage with their preexisting worldviews and specific topics of concern, and 2) the difference between moderate and far-right publishers when we examine the writing style of their articles. Furthermore, we can identify users prone to sharing misinformation based on their communication style. These findings have important implications for countering online misinformation, as they highlight the powerful role that personal biases towards specific topics and publishers' writing styles have in amplifying fringe ideologies online.

cs.CL↗

arXiv2024

Harmony in the Australian Domain Space

In this paper we use for the first time a systematic approach in the study of harmonic centrality at a Web domain level, and gather a number of significant new findings about the Australian web. In particular, we explore the relationship between economic diversity at the firm level and the structure of the Web within the Australian domain space, using harmonic centrality as the main structural feature. The distribution of harmonic centrality values is analyzed over time, and we find that the distributions exhibit a consistent pattern across the different years. The observed distribution is well captured by a partition of the domain space into six clusters; the temporal movement of domain names across these six positions yields insights into the Australian Domain Space and exhibits correlations with other non-structural characteristics. From a more global perspective, we find a significant correlation between the median harmonic centrality of all domains in each OECD country and one measure of global trust, the WJP Rule of Law Index. Further investigation demonstrates that 35 countries in OECD share similar harmonic centrality distributions. The observed homogeneity in distribution presents a compelling avenue for exploration, potentially unveiling critical corporate, regional, or national insights.

physics.soc-ph↗

arXiv2024

Opinion Market Model: Stemming Far-Right Opinion Spread using Positive Interventions

Online extremism has severe societal consequences, including normalizing hate speech, user radicalization, and increased social divisions. Various mitigation strategies have been explored to address these consequences. One such strategy uses positive interventions: controlled signals that add attention to the opinion ecosystem to boost certain opinions. To evaluate the effectiveness of positive interventions, we introduce the Opinion Market Model (OMM), a two-tier online opinion ecosystem model that considers both inter-opinion interactions and the role of positive interventions. The size of the opinion attention market is modeled in the first tier using the multivariate discrete-time Hawkes process; in the second tier, opinions cooperate and compete for market share, given limited attention using the market share attraction model. We demonstrate the convergence of our proposed estimation scheme on a synthetic dataset. Next, we test OMM on two learning tasks, applying to two real-world datasets to predict attention market shares and uncover latent relationships between online items. The first dataset comprises Facebook and Twitter discussions containing moderate and far-right opinions about bushfires and climate change. The second dataset captures popular VEVO artists' YouTube and Twitter attention volumes. OMM outperforms the state-of-the-art predictive models on both datasets and captures latent cooperation-competition relations. We uncover (1) self- and cross-reinforcement between far-right and moderate opinions on the bushfires and (2) pairwise artist relations that correlate with real-world interactions such as collaborations and long-lasting feuds. Lastly, we use OMM as a testbed for positive interventions and show how media coverage modulates the spread of far-right opinions.

cs.SI↗

arXiv2023

Transfer Learning for Hate Speech Detection in Social Media

Today, the internet is an integral part of our daily lives, enabling people to be more connected than ever before. However, this greater connectivity and access to information increase exposure to harmful content such as cyber-bullying and cyber-hatred. Models based on machine learning and natural language offer a way to make online platforms safer by identifying hate speech in web text autonomously. However, the main difficulty is annotating a sufficiently large number of examples to train these models. This paper uses a transfer learning technique to leverage two independent datasets jointly and builds a single representation of hate speech. We build an interpretable two-dimensional visualization tool of the constructed hate speech representation -- dubbed the Map of Hate -- in which multiple datasets can be projected and comparatively analyzed. The hateful content is annotated differently across the two datasets (racist and sexist in one dataset, hateful and offensive in another). However, the common representation successfully projects the harmless class of both datasets into the same space and can be used to uncover labeling errors (false positives). We also show that the joint representation boosts prediction performances when only a limited amount of supervision is available. These methods and insights hold the potential for safer social media and reduce the need to expose human moderators and annotators to distressing online messaging.

cs.SI↗

arXiv2023

The Innovation-to-Occupations Ontology: Linking Business Transformation Initiatives to Occupations and Skills

The fast adoption of new technologies forces companies to continuously adapt their operations making it harder to predict workforce requirements. Several recent studies have attempted to predict the emergence of new roles and skills in the labour market from online job ads. This paper aims to present a novel ontology linking business transformation initiatives to occupations and an approach to automatically populating it by leveraging embeddings extracted from job ads and Wikipedia pages on business transformation and emerging technologies topics. To our knowledge, no previous research explicitly links business transformation initiatives, like the adoption of new technologies or the entry into new markets, to the roles needed. Our approach successfully matches occupations to transformation initiatives under ten different scenarios, five linked to technology adoption and five related to business. This framework presents an innovative approach to guide enterprises and educational institutions on the workforce requirements for specific business transformation initiatives.

cs.AI↗

arXiv2023

Enhancing Biomedical Text Summarization and Question-Answering: On the Utility of Domain-Specific Pre-Training

Biomedical summarization requires large datasets to train for text generation. We show that while transfer learning offers a viable option for addressing this challenge, an in-domain pre-training does not always offer advantages in a BioASQ summarization task. We identify a suitable model architecture and use it to show a benefit of a general-domain pre-training followed by a task-specific fine-tuning in the context of a BioASQ summarization task, leading to a novel three-step fine-tuning approach that works with only a thousand in-domain examples. Our results indicate that a Large Language Model without domain-specific pre-training can have a significant edge in some domain-specific biomedical text generation tasks.

cs.CL↗