Accurate time-series forecasting for complex physical systems is the backbone of modern industrial monitoring and control. While deep learning models excel at capturing complex dynamics, currently, their deployment is limited due to physical inconsistency and robustness, hence constraining their reliability in regulated environments. We introduce process-informed forecasting (PIF) models for temperature in pharmaceutical lyophilization. We investigate a wide range of models, from classical ones such as Autoregressive Integrated Moving Average Model (ARIMA) and Exponential Smoothing Model (ETS), to modern deep learning architectures, including Kolmogorov-Arnold Networks (KANs). We compare three different loss function formulations that integrate a process-informed trajectory prior: a fixed-weight loss, a dynamic uncertainty-based loss, and a Residual-Based Attention (RBA) mechanism. We evaluate all models not only for accuracy and physical consistency but also for robustness to sensor noise. Furthermore, we test the practical generalizability of the best model in a transfer learning scenario on a new process. Our results show that PIF models outperform their data-driven counterparts in terms of accuracy, physical plausibility and noise resilience. This work provides a roadmap for developing reliable and generalizable forecasting solutions for critical applications in the pharmaceutical manufacturing landscape.

Account Takeover (ATO) fraud poses a significant challenge in consumer banking, requiring high recall under strict latency while minimizing friction for legitimate users. Production systems typically rely on tabular gradient-boosted decision trees (e.g., XGBoost) that score sessions independently, overlooking the relational and temporal structure of online activity that characterizes coordinated attacks and "fraud rings." We introduce ATLAS (Account Takeover Learning Across Spatio-Temporal Directed Graph), a framework that reformulates ATO detection as spatio-temporal node classification on a time-respecting directed session graph. ATLAS links entities via shared identifiers (account, device, IP) and regulates connectivity with time-window and recency constraints, enabling causal, time-respecting message passing and latency-aware label propagation that uses only labels available at scoring time, non-anticipative and leakage-free. We operationalize ATLAS with inductive GraphSAGE variants trained via neighbor sampling, at scale on a sessions graph with more than 100M nodes and around 1B edges. On a high-risk digital product at Capital One, ATLAS delivers 6.38 percent AUC improvement and more than 50 percent reduction in customer friction, improving fraud capture while reducing user friction.

--Typical financial risk management involves distinct phases for pre-service risk assessment and in-service default detection, often modeled separately. This paper proposes a novel framework, Multi-Granularity Knowledge Distillation (abbreviated as MGKD), aimed at improving pre-service risk prediction through the integration of in-service user behavior data. MGKD follows the idea of knowledge distillation, where the teacher model, trained on historical in-service data, guides the student model, which is trained on pre-service data. By using soft labels derived from in-service data, the teacher model helps the student model improve its risk prediction prior to service activation. Meanwhile, a multi-granularity distillation strategy is introduced, including coarse-grained, fine-grained, and self-distillation, to align the representations and predictions of the teacher and student models. This approach not only reinforces the representation of default cases but also enables the transfer of key behavioral patterns associated with defaulters from the teacher to the student model, thereby improving the overall performance of pre-service risk assessment. Moreover, we adopt a re-weighting strategy to mitigate the model's bias towards the minority class. Experimental results on large-scale real-world datasets from T encent Mobile Payment demonstrate the effectiveness of our proposed approach in both offline and online scenarios. Risk management has been central to financial services, permeating every stage of their lifecycle.

Traffic accidents remain a major global concern, with lane-change maneuvers recognized as one of the significant contributors to collision risk. Anticipating these maneuvers has become an important research focus, supporting both traffic safety and the safe integration of autonomous and assisted driving technologies. Over the past decade, numerous models have been developed for lane-change prediction. However, most existing works have been designed and validated using simulation environments or pre-recorded datasets. While these settings allow for benchmarking and controlled evaluation, they often rely on simplified assumptions about sensing, communication, and vehicle behavior that do not fully capture the complexity of real-world operation. Real-world deployments of lane-change prediction systems are relatively rare, and when they are reported, their practical challenges, limitations, and insights remain under-documented. To illustrate the setting more concretely, consider the left lane change scenario shown in Figure 1. The Ego Vehicle (EV) is driving in the left lane, while the Target Vehicle (TV) is moving in the right lane behind a Preceding Vehicle (PV). When the PV suddenly brakes, the TV must change lanes to avoid a collision.

Model inversion attacks (MIAs) aim to reconstruct class-representative samples from trained models. Recent generative MIAs utilize generative adversarial networks to learn image priors that guide the inversion process, yielding reconstructions with high visual quality and strong fidelity to the private training data. To explore the reason behind their effectiveness, we begin by examining the gradients of inversion loss with respect to synthetic inputs, and find that these gradients are surprisingly noisy. Further analysis reveals that generative inversion implicitly denoises these gradients by projecting them onto the tangent space of the generator manifold, filtering out off-manifold components while preserving informative directions aligned with the manifold. Our empirical measurements show that, in models trained with standard supervision, loss gradients often exhibit large angular deviations from the data manifold, indicating poor alignment with class-relevant directions. This observation motivates our central hypothesis: models become more vulnerable to MIAs when their loss gradients align more closely with the generator manifold. We validate this hypothesis by designing a novel training objective that explicitly promotes such alignment. Building on this insight, we further introduce a training-free approach to enhance gradient-manifold alignment during inversion, leading to consistent improvements over state-of-the-art generative MIAs.

The "No Free Lunch" theorem dictates that no single recommender algorithm is optimal for all users, creating a significant Algorithm Selection Problem. Standard meta-learning approaches aim to solve this by selecting an algorithm based on user features, but treat the fundamentally diverse algorithms themselves as equivalent, "black-box" choices. This thesis investigates the impact of overcoming this limitation by engineering a comprehensive feature set to explicitly characterize the algorithms themselves. We combine static code metrics, Abstract Syntax Tree properties, behavioral performance landmarks, and high-level conceptual features. We evaluate two meta-learners across five datasets: a baseline using only user features and our proposed model using both user and algorithm features. Our results show that the meta-learner augmented with algorithm features achieves an average NDCG@10 of 0.143, a statistically significant improvement of 11.7% over the Single Best Algorithm baseline (0.128). However, we found that the inclusion of algorithm features did not lead to an improvement in overall NDCG@10 over the meta learner using only user features (0.144). While adding algorithm features to the meta-learner did improve its Top-1 selection accuracy (+16.1%), this was counterbalanced by leading to a lower Top-3 accuracy (-10.7%). We conclude that for the per-user algorithm selection task in recommender systems, the predictive power of user features is overwhelmingly dominant. While algorithm features improve selection precision, unlocking their potential to boost overall performance remains a non-trivial challenge.

Language models often struggle with idiomatic, figurative, or context-sensitive inputs, not because they produce flawed outputs, but because they misinterpret the input from the outset. We propose an input-only method for anticipating such failures using token-level likelihood features inspired by surprisal and the Uniform Information Density hypothesis. These features capture localized uncertainty in input comprehension and outperform standard baselines across five linguistically challenging datasets. We show that span-localized features improve error detection for larger models, while smaller models benefit from global patterns. Our method requires no access to outputs or hidden activations, offering a lightweight and generalizable approach to pre-generation error prediction.

Robotic manipulation tasks involving cutting deformable objects remain challenging due to complex topological behaviors, difficulties in perceiving dense object states, and the lack of efficient evaluation methods for cutting outcomes. In this paper, we introduce TopoCut, a comprehensive benchmark for multi-step robotic cutting tasks that integrates a cutting environment and generalized policy learning. TopoCut is built upon three core components: (1) We introduce a high-fidelity simulation environment based on a particle-based elastoplastic solver with compliant von Mises constitutive models, augmented by a novel damage-driven topology discovery mechanism that enables accurate tracking of multiple cutting pieces. (2) We develop a comprehensive reward design that integrates the topology discovery with a pose-invariant spectral reward model based on Laplace-Beltrami eigenanalysis, facilitating consistent and robust assessment of cutting quality. (3) We propose an integrated policy learning pipeline, where a dynamics-informed perception module predicts topological evolution and produces particle-wise, topology-aware embeddings to support PDDP (Particle-based Score-Entropy Discrete Diffusion Policy) for goal-conditioned policy learning. Extensive experiments demonstrate that TopoCut supports trajectory generation, scalable learning, precise evaluation, and strong generalization across diverse object geometries, scales, poses, and cutting goals.

Positional encoding is essential for supplementing transformer with positional information of tokens. Existing positional encoding methods demand predefined token/feature order, rendering them unsuitable for real-world data with non-sequential yet causally-related features. To address this limitation, we propose CAPE, a novel method that identifies underlying causal structure over non-sequential features as a weighted directed acyclic graph (DAG) using generalized structural equation modeling. The DAG is then embedded in hyperbolic space where its geometric structure is well-preserved using a hyperboloid model-based approach that effectively captures two important causal graph properties (causal strength & causal specificity). This step yields causality-aware positional encodings for the features, which are converted into their rotary form for integrating with transformer's self-attention mechanism. Theoretical analysis reveals that CAPE-generated rotary positional encodings possess three valuable properties for enhanced self-attention, including causal distance-induced attenuation, causal generality-induced attenuation, and robustness to positional disturbances. We evaluate CAPE over both synthetic and real-word datasets, empirically demonstrating its theoretical properties and effectiveness in enhancing transformer for data with non-sequential features. Our code is available at https://github.com/Catchxu/CAPE.

Ensuring reliability is paramount in deep learning, particularly within the domain of medical imaging, where diagnostic decisions often hinge on model outputs. The capacity to separate out-of-distribution (OOD) samples has proven to be a valuable indicator of a model's reliability in research. In medical imaging, this is especially critical, as identifying OOD inputs can help flag potential anomalies that might otherwise go undetected. While many OOD detection methods rely on feature or logit space representations, recent works suggest these approaches may not fully capture OOD diversity. To address this, we propose a novel OOD scoring mechanism, called NERO, that leverages neuron-level relevance at the feature layer. Specifically, we cluster neuron-level relevance for each in-distribution (ID) class to form representative centroids and introduce a relevance distance metric to quantify a new sample's deviation from these centroids, enhancing OOD separability. Additionally, we refine performance by incorporating scaled relevance in the bias term and combining feature norms. Our framework also enables explainable OOD detection.