Researchers have developed ACFormer, a novel approach that blends convolutional efficiency with the power of attention mechanisms to dramatically improve time series prediction.
A novel forecasting framework leverages the power of generative models to anticipate day-ahead electricity market curves, enabling smarter energy storage strategies.
![Explanation-guided graph neural networks, despite achieving accurate predictions, are susceptible to generating ‘degenerate’ explanations - outputs that correctly encode the predicted label by exploiting irrelevant, consistently-colored nodes - effectively misleading users into believing these nodes drive the model’s inference when they do not, a phenomenon demonstrated by the network’s ability to utilize an ‘anchor set’ [latex]\mathcal{Z}[/latex] of class-indiscriminative nodes to secretly encode predictions.](https://arxiv.org/html/2601.20815v1/figs/violet.png)
Despite achieving high accuracy, many self-explainable graph neural networks offer explanations that don’t reflect how they actually make decisions, raising concerns about their trustworthiness.
A new study explores how effectively we can train AI to identify content generated by other AI systems.
A new system leverages the power of large language models to deliver personalized and actionable insights for online sellers.
![The simulation demonstrates that market valuations, even when driven by as few as 100 agents over 100 time steps, consistently reflect underlying election outcomes, though predictive accuracy is predictably degraded by individual biases, risk aversion, and the inherent noise within both the market price and the true result-particularly around a 0.5 probability, where misclassification risk increases and confidence in alignment between prediction and reality diminishes, as evidenced by the widening transition zone between correct and incorrect classifications with increased uncertainty, despite parameter variations tested across expertise, stubbornness, and budget constraints [latex] B\_{i,0}\sim U(100,1000), V\_{i,0}\sim N(0.5,0.05), s\_{i}\sim\mathcal{N}(0.3,0.05), e\_{i}\sim\mathcal{N}(0.9,0.04), r\_{i}\sim U(0,1) [/latex].](https://arxiv.org/html/2601.20452v1/x1.png)
New research reveals that even generally reliable prediction markets are vulnerable to price manipulation by large, well-funded participants, especially when combined with investor herd mentality.
![The ARMAConv Encoder-Only Transformer (ACEOT) architecture processes weighted adjacency matrices, node-level power injections [latex] (P, Q) [/latex], and node indices to simultaneously estimate node-level attack probabilities for precise localization and a comprehensive graph-level probability for False Data Injection Attacks (FDIA) detection.](https://arxiv.org/html/2601.18981v1/x6.png)
A new artificial intelligence framework leverages advanced graph neural networks to dramatically improve the detection and localization of false data injection attacks targeting power grids.
New research establishes a formal link between the architectural depth of convolutional networks and their ability to efficiently learn hierarchical representations of data.
![The study demonstrates that identifying subgroups experiencing the greatest unfairness-particularly within conjunction subgroups [latex]\mathcal{S}_A[/latex] as a subset of broader linear subgroups [latex]\mathcal{S}_L[/latex]-yields optimal solutions, as evidenced by the performance of MSD and GerryFair relative to alternative approaches.](https://arxiv.org/html/2601.19595v1/x1.png)
New research tackles the complex challenge of intersectional bias in machine learning models, offering a robust framework for fairer and more equitable results.
A new approach leverages immediate feedback and informed constraints to dramatically improve policy exploitation in online reinforcement learning environments.