Humans and large language models (LLMs) now co-produce and co-consume the web's shared knowledge archives. Such human-AI collective knowledge ecosystems contain feedback loops with both benefits (e.g., faster growth, easier learning) and systemic risks (e.g., quality dilution, skill reduction, model collapse). To understand such phenomena, we propose a minimal, interpretable dynamical model of the co-evolution of archive size, archive quality, model (LLM) skill, aggregate human skill, and query volume. The model captures two content inflows (human, LLM) controlled by a gate on LLM-content admissions, two learning pathways for humans (archive study vs. LLM assistance), and two LLM-training modalities (corpus-driven scaling vs. learning from human feedback). Through numerical experiments, we identify different growth regimes (e.g., healthy growth, inverted flow, inverted learning, oscillations), and show how platform and policy levers (gate strictness, LLM training, human learning pathways) shift the system across regime boundaries. Two domain configurations (PubMed, GitHub and Copilot) illustrate contrasting steady states under different growth rates and moderation norms. We also fit the model to Wikipedia's knowledge flow during pre-ChatGPT and post-ChatGPT eras separately. We find a rise in LLM additions with a concurrent decline in human inflow, consistent with a regime identified by the model. Our model and analysis yield actionable insights for sustainable growth of human-AI collective knowledge on the Web.

Online social networks have greatly facilitated the spread of information of all sorts. Meanwhile, the abundance of information in today's world also means different pieces of information will increasingly compete for people's finite attention. When different pieces of information spread together in an online social network, why would some become trendy while others fail to emerge? Existing research either models the diffusion of each piece of information independently, or fails to consider users' inactivity in online social networks. Modeling each piece of information as a meme, this paper addresses this gap by proposing a unified model for the co-diffusion of competing memes simultaneously spreading across an online social network. We are the first to identify a ubiquitous threshold for competing meme. The threshold also functions as an effective predictor that contributes to better performance in determining the outcome of meme competitions. Outcomes from this study have important implications for online campaigns and mobilizations as well as the fight against misinformation. • Unified model for competing memes' spread in online networks. • First to identify a universal meme competition threshold. • Threshold predicts meme competition outcomes effectively.

Graph representation learning is an emerging area for graph analysis and inference. However, existing approaches for large-scale graphs either sample nodes in sequential walks or manipulate the adjacency matrices of graphs. The former approach can cause sampling bias against less-connected nodes, whereas the latter may suffer from sparsity that exists in many real-world graphs. To learn from structural information in a graph more efficiently and comprehensively, this paper proposes a new graph representation learning approach inspired by the cognitive model of spreading-activation mechanisms in human memory. This approach learns node embeddings by adopting a graph activation model that allows nodes to “activate” their neighbors and spread their own structural information to other nodes through the paths simultaneously. Comprehensive experiments demonstrate that the proposed model performs better than existing methods on several empirical datasets for multiple graph inference tasks. Meanwhile, the spreading-activation-based model is computationally more efficient than existing approaches–the training process converges after only a small number of iterations, and the training time is linear in the number of edges in a graph. The proposed method works for both homogeneous and heterogeneous graphs.

Co-movement among individual firms’ stock prices can reflect complex interfirm relationships. This paper proposes a novel method to leverage such relationships for stock price predictions by adopting inductive graph representation learning on dynamic stock graphs constructed based on historical stock price co-movement. To learn node representations from such dynamic graphs for better stock predictions, we propose the hybrid-attention dynamic graph neural network, an inductive graph representation learning method. We also extended mini-batch gradient descent to inductive representation learning on dynamic stock graphs so that the model can update parameters over mini-batch stock graphs with higher training efficiency. Extensive experiments on stocks from different markets and trading simulations demonstrate that the proposed method significantly improves stock predictions. The proposed method can have important implications for the management of financial portfolios and investment risk. Summary of Contribution: Accurate predictions of stock prices have important implications for financial decisions. In today’s economy, individual firms are increasingly connected via different types of relationships. As a result, firms’ stock prices often feature synchronous co-movement patterns. This paper represents the first effort to leverage such phenomena to construct dynamic stock graphs for stock predictions. We develop hybrid-attention dynamic graph neural network (HAD-GNN), an inductive graph representation learning framework for dynamic stock graphs to incorporate temporal and graph attention mechanisms. To improve the learning efficiency of HAD-GNN, we also extend the mini-batch gradient descent to inductive representation learning on such dynamic graphs and adopt a t-batch training mechanism (t-BTM). We demonstrate the effectiveness of our new approach via experiments based on real-world data and simulations.

The spread of misinformation on social media has become a major societal issue during recent years. In this work, we used the ongoing COVID-19 pandemic as a case study to systematically investigate factors associated with the spread of multi-topic misinformation related to one event on social media based on the heuristic-systematic model. Among factors related to systematic processing of information, we discovered that the topics of a misinformation story matter, with conspiracy theories being the most likely to be retweeted. As for factors related to heuristic processing of information, such as when citizens look up to their leaders during such a crisis, our results demonstrated that behaviors of a political leader, former US President Donald J. Trump, may have nudged people's sharing of COVID-19 misinformation. Outcomes of this study help social media platform and users better understand and prevent the spread of misinformation on social media.

Abstract Online health communities (OHCs) have been major resources for people with similar health concerns to interact with each other. They offer easily accessible platforms for users to seek, receive, and provide supports by posting. Taking the advantage of text mining and machine learning techniques, we identified social support type(s) in each post and a new user's support needs in an OHC. We examined a user's first‐time support‐seeking experience by measuring both quantity and quality of received support. Our results revealed that the amount and match of received support are positive and significant predictors of new users' continued engagement. Our outcomes can provide insight for designing and managing a sustainable OHC by retaining users.

Abstract Performance assessment is ubiquitous and crucial in people analytics. Scientific impact, particularly, plays a significant role in the academia. This paper attempts to understand researchers' career trajectories by considering the research community as a social network, where individuals build ties with each other via coauthorship. The resulting linkage facilitates information flow and affects researchers' future impact. Consequently, we systematically investigate the career trajectories of researchers with respect to research impact using the social capital theory as our theoretical foundation. Specifically, for early‐stage and mid‐career academics, we find that connections with prominent researchers associate with greater impact. Brokerage positions, in addition, are beneficial to a researcher's research impact in the long run. For senior researchers, however, the only social network feature that significantly affects their future impact is the reputation of their recently built ties. Finally, we build predictive models on future research impact which can be leveraged by both organizations and individuals. This paper provides empirical evidence for how social networks provide signals on researchers' career dynamics guided by social capital theory. Our findings have implications for individual researchers to strategically plan and promote their careers and for research institutions to better evaluate current as well as prospective employees.

The past 2 decades have witnessed the emergence of information as a scientific discipline and the growth of information schools around the world. We analyzed the current state of the iSchool community in the U.S. with a special focus on the evolution of the community. We conducted our study from the perspectives of acquiring talents and producing research, including the analysis on iSchool faculty members' educational backgrounds, research topics, and the hiring network among iSchools. Applying text mining techniques and social network analysis to data from various sources, our research revealed how the iSchool community gradually built its own identity over time, including the growing number of faculty members who received their doctorates from the field that studies information, the deviation from computer science and library science, the rising emphasis on the intersection of information, technology, and people, and the increasing educational and research homogeneity as a community. These findings suggest that iSchools in the U.S. are evolving into a mature and independent discipline with a more established identity.