Association for the Advancement of Artificial Intelligence: AAAI Publications
Not a member yet
26155 research outputs found
Sort by
A Benchmark for Cross-Domain Argumentative Stance Classification on Social Media
No full textArgumentative stance classification plays a key role in identifying authors' viewpoints on specific topics. However, generating diverse pairs of argumentative sentences across various domains is challenging. Existing benchmarks often come from a single domain or focus on a limited set of topics. Additionally, manual annotation for accurate labeling is time-consuming and labor-intensive. To address these challenges, we propose leveraging platform rules, readily available expert-curated content, and large language models to bypass the need for human annotation. Our approach produces a multidomain benchmark comprising 4,498 topical claims and 30,961 arguments from three sources, spanning 21 domains. We benchmark the dataset in fully supervised, zero-shot, and few-shot settings, shedding light on the strengths and limitations of different methodologies
Paths of A Million People: Extracting Life Trajectories from Wikipedia
No full textThe life trajectories of notable people have been studied to pinpoint the times and places of significant events such as birth, death, education, marriage, competition, work, speeches, scientific discoveries, artistic achievements, and battles. Understanding how these individuals interact with others provides valuable insights for broader research into human dynamics.
However, the scarcity of trajectory data in terms of volume, density, and inter-person interactions, limits relevant studies from being comprehensive and interactive. We mine millions of biography pages from Wikipedia and tackle the generalization problem stemming from the variety and heterogeneity of the trajectory descriptions. Our ensemble model COSMOS, which combines the idea of semi-supervised learning and contrastive learning, achieves an F1 score of 85.95%. For this task, we also create a hand-curated dataset, WikiLifeTrajectory, consisting of 8,852 (person, time, location) triplets as ground truth. Besides, we perform an empirical analysis on the trajectories of 8,272 historians to demonstrate the validity of the extracted results. To facilitate the research on trajectory extractions and help the analytical studies to construct grand narratives, we make our code, the million-level extracted trajectories, and the WikiLifeTrajectory dataset publicly available
Quantifying the Spread of Online Incivility in Brazilian Politics
No full textIncivility refers to behaviors that violate collective norms and disrupt cooperation within the political process. Although large-scale online data and automated techniques have enabled the quantitative analysis of uncivil discourse, prior research has predominantly focused on impoliteness or toxicity, often overlooking other behaviors that undermine democratic values. To address this gap, we propose a multidimensional conceptual framework encompassing Impoliteness (IMP), Physical Harm and Violent Political Rhetoric (PHAVPR), Hate Speech and Stereotyping (HSST), and Threats to Democratic Institutions and Values (THREAT). Using this framework, we measure the spread of online political incivility in Brazil using approximately 5 million tweets posted by 2,307 political influencers during the 2022 Brazilian general election. Through statistical modeling and network analysis, we examine the dynamics of uncivil posts at different election stages, identify key disseminators and audiences, and explore the mechanisms driving the spread of uncivil information online. Our findings indicate that impoliteness is more likely to surge during election campaigns. In contrast, the other dimensions of incivility are often triggered by specific violent events. Moreover, we find that left-aligned individual influencers are the primary disseminators of online incivility in the Brazilian Twitter/X sphere and that they disseminate not only direct incivility but also indirect incivility when discussing or opposing incivility expressed by others. They relay those content from politicians, media agents, and individuals to reach broader audiences, revealing a diffusion pattern mixing the direct and two-step flows of communication theory. This study offers new insights into the multidimensional nature of incivility in Brazilian politics and provides a conceptual framework that can be extended to other political contexts
SAID: A Social Media AI-generated Interface Dataset Using Prompt Engineering Methods Focused On Accessibility
No full textWe present SAID (Social Media AI-generated Interface Dataset), a systematically curated collection of 240 social media profile interfaces generated through controlled prompt engineering focused on accessibility. As AI tools reshape interface design practices, understanding how these systems interpret and implement accessibility requirements in social media interfaces becomes more and more important. Through six distinct prompt categories examining both generic and specific accessibility requirements, our dataset captures how AI systems interpret and implement accessibility features across visual and motor impairment dimensions. The dataset combines complete interface designs in multiple formats (PNG and SVG), detailed prompt engineering methodology, and comprehensive documentation of interface components such as social identity presentation, content engagement, navigation, and interactive elements. SAID enables novel research directions from understanding AI's role in shaping accessible social media experiences to examining how automated design tools can support more inclusive social interactions
Mobility Networked Time-Series Forecasting Benchmark Datasets
No full textHuman mobility is crucial for urban planning (e.g., public transportation) and epidemic response strategies. However, existing research often neglects integrating comprehensive perspectives on spatial dynamics, temporal trends, and other contextual views due to the limitations of existing mobility datasets. To bridge this gap, we introduce MOBINS (MOBIlity Networked time Series), a novel dataset collection designed for networked time-series forecasting of dynamic human movements. MOBINS features diverse and explainable datasets that capture various mobility patterns across different transportation modes in four cities and two countries and cover both transportation and epidemic domains at the administrative area level. Our experiments with nine baseline methods reveal the significant impact of different model backbones on the proposed six datasets. We provide a valuable resource for advancing urban mobility research
Labeled Datasets for Research on Information Operations
No full textSocial media platforms have become a hub for political activities and discussions, democratizing participation in these endeavors. However, they have also become an incubator for manipulation campaigns, like information operations (IOs). Some social media platforms have released datasets related to such IOs originating from different countries. However, we lack comprehensive control data that can enable the development of IO detection methods. To bridge this gap, we present new labeled datasets about 26 campaigns, which contain both IO posts verified by a social media platform and over 13M posts by 303k accounts that discussed similar topics in the same time frames (control data). The datasets will facilitate the study of narratives, network interactions, and engagement strategies employed by coordinated accounts across various campaigns and countries. By comparing these coordinated accounts against organic ones, researchers can develop and benchmark IO detection algorithms
ThatiAR: Subjectivity Detection in Arabic News Sentences
No full textIn this study, we present the first large dataset, ThatiAR, for subjectivity detection in Arabic, consisting of ~3.6K manually annotated sentences, and GPT-4o based explanations. In addition, we include instructions (both in English and Arabic) to facilitate LLM based fine-tuning. We provide an in-depth analysis of the dataset, annotation process, and extensive benchmark results, including PLMs and LLMs. Our analysis of the annotation process highlights that annotators were strongly influenced by their political, cultural, and religious backgrounds, especially at the beginning of the annotation process. The experimental results suggest that LLMs with in-context learning provide better performance. We release the dataset and resources to the community
Co-HSF: Resource-Efficient One-Shot Semi-Supervised Adaptation of Histopathology Foundation Models
No full textAutomated analysis of histopathological samples has greatly augmented the ability of experts to perform deep phenotyping on biological samples. Current state-of-the-art (SOTA) methods for histopathology image classification rely on training deep neural networks with large annotated datasets, which can be costly to obtain. Recent studies propose to bypass annotated datasets by leveraging pre-trained foundation models (e.g. visual-language models) for zero-shot predictions. Moreover, fine-tuning these models enhances performance while requiring minimal labeled data (e.g. one-shot fine-tuning). However, one-shot fine-tuned performance of histopathology foundation models on image classification tasks is understudied. In this work, we first explore the use of semi-supervised few-shot learning (SSFSL) for fine-tuning histopathology foundation models on one-shot datasets with unlabeled samples. We find SOTA SSFSL methods improve fine-tuning performance, but their pseudo-labeling (i.e. assigning labels to unlabeled samples) strategies can increase inference times over zero-shot. We then propose a Co-filtered Histopathology Semi-Supervised Few-Shot (Co-HSF) pipeline: a dual-SSFSL (i.e. with teacher and student models) training loop followed by a co-filtering (CF) pseudo-labeling strategy to efficiently leverage unlabeled data for improved semi-supervised performance and reduced inference times. Using the National Center for Tumor Disease Colorectal Cancer Dataset (NCT-CRC-HE), we show our proposed module achieves 38.4% improvement in accuracy over zero-shot performance with only 9 labeled samples and over 53% faster inference times, while also outperforming other fine-tuning and SSFSL methods
Beyond Labels: A Self-Supervised Framework with Masked Autoencoders and Random Cropping for Breast Cancer Sub-type Classification
No full textThis work addresses the problem of breast cancer sub-type classification using histopathological image analysis. We utilize masked autoencoders (MAEs) based on Visual Transformer (ViT) to learn, through Self-Supervised Learning, embeddings tailored to computer vision tasks in this domain. Such embeddings capture informative representations of histopathological data, facilitating feature learning without extensive labeled datasets. During pre-training, we investigate employing a random crop technique to generate a large dataset from whole-slide images automatically. Additionally, we assess the performance of linear probes for multi-class classification tasks of cancer sub-types using the representations learned by the MAE. Our approach aims to achieve strong performance on downstream tasks by leveraging the complementary strengths of ViTs and autoencoders. We evaluate our model's performance on the BRACS and BACH datasets and compare it with existing benchmarks
Human-AI Collaborations for Controlled Tasking Use Cases
No full textMany Human-AI collaborations are based on task environments in which tasks on controllable assets need to be taken (e.g., to change a route of a truck or to send an informational message to the truck driver) in response to results of monitoring tasks (e.g., changes in customer needs, changes in weather). Human-AI teams need to collaborate effectively and efficiently to generate, assess, and adjust Courses of Actions (COAs) in those environments.
RTX BBN Technologies (BBN) has developed multiple AI agents, covering different specific use cases, and is currently deploying Human-AI teams into operational environments. In our work, the most relevant use cases involve an AI agent advising a human on executing complex tasks and human and AI agents handling subtasks usually done by one person.
The workshop presentation will focus on overall context, our successes with Human-AI teams in recent years, and specific insights from our work with Human-AI teams performing joint activities in controlled tasking environments