Many businesses depend on their mobile apps and websites, so user frustration while trying to complete a task on these channels can cause lost sales and complaints. In this research, I use clickstream data from a real e-commerce site to predict whether a session is frustrated or not. Frustration is defined using certain rules based on rage bursts, back and forth navigation (U turns), cart churn, search struggle, and long wandering sessions, and applies these rules to 5.4 million raw clickstream events (304,881 sessions). From each session, I build tabular features and train standard classifier models. I also use the full event sequence to train a discriminative LSTM classifier. XGBoost reaches about 90% accuracy, ROC AUC of 0.9579, while the LSTM performs best with about 91% accuracy and a ROC AUC of 0.9705. Finally, the research shows that with only the first 20 to 30 interactions, the LSTM already predicts frustration reliably.
Subsurface delaminations in reinforced concrete bridge decks escape conventional visual inspection, and the two principal sensing techniques used to find them are individually incomplete: Ground Penetrating Radar (GPR) penetrates deeply but degrades near the surface, while Infrared Thermography (IRT) resolves shallow defects but cannot reach deeper structure. This paper presents a framework for fusing the two modalities through hierarchical attention: temporal self-attention over GPR A-scans, channel-spatial attention over IRT patches, and cross-modal multi-head attention with learnable modality embeddings, coupled with decomposed aleatoric/epistemic uncertainty estimation. Beyond the architecture itself, which is lightweight at approximately 0.53M parameters with a closed-form accounting of where capacity resides, we contribute an elementary formal analysis. Two-token cross-modal attention is shown to be exactly a bank of per-sample learned gates; a gradient-allocation proposition quantifies how class imbalance starves attention parameters of minority-class signal and how loss reweighting trades that starvation for gradient variance; and closed-form metric floors under majority-class collapse anchor a diagnostic divergence between ranking metrics (AUC) and thresholded metrics (F1). The analysis suggests that adaptively weighted fusion, precisely because its feature-selection policy is learned, may be distinctively vulnerable to the severe class imbalance typical of operational bridge decks; establishing whether and when this occurs is deferred to empirical evaluation.
Biological systems can form complex three-dimensional structures through the collective behavior of agents that share a common update rule and operate without central control. How such distributed control gives rise to precise global patterns remains a central question not only in developmental biology but also in distributed robotics, programmable matter, and multi-agent learning. Here, we introduce DiffeoMorph, an end-to-end differentiable framework for learning a morphogenesis protocol that guides a population of agents to morph into a target 3D shape. Each agent updates its position and internal state using an SE(3)-equivariant graph neural network, based on its own internal state and signals received from other agents. To train this system, we introduce a new shape-matching loss based on 3D Zernike polynomials, which compares the predicted and target shapes as continuous spatial distributions, not as discrete point clouds, and is invariant to agent ordering, number of agents, and global orientation. To achieve rotation invariance while preserving reflection sensitivity, we include an alignment step that optimally rotates the predicted Zernike spectrum to match the target before computing the loss. We perform benchmarking to establish the advantages of our shape-matching loss over other standard distance metrics for shape comparison tasks. We then demonstrate that DiffeoMorph can form a range of complex shapes from minimally patterned initial conditions. DiffeoMorph provides a general framework for learning distributed control strategies for morphogenesis, swarm robotics, and programmable self-assembly.
In this work, we propose a method to automatically find the regularization parameter for low-rank MMSE filters based on a Kronecker-product representation, using an approximate leave-one-out (ALO) cross-validation metric. We formulate the problem in a constrained form that covers both error minimization and interference suppression, and explore the link between regularization and rank selection. The proposed method is validated on system identification and antenna beamforming problems, showing significant gains over commonly used methods.
End-to-end vision-based controllers map raw camera images directly to control actions, yet verifying their closed-loop behavior remains challenging: no analytical model maps a physical state to a camera image, and propagating a set of states through high-dimensional images yields loose reachable sets. We address both challenges with two techniques. First, we propose a Deterministic World Model (DWM), a latent-free decoder that generates camera images only from the physical state, trained with a saliency-weighted reconstruction loss and a control-consistency loss. The DWM replaces the camera in the closed loop, so a sound and tight reachable tube of the surrogate closed loop can be computed without bounding any latent variables. Second, we propose an action-guided partition-and-verify scheme that certifies as many initial sets as possible within a fixed computation budget. A Gaussian process learned from cheap simulations orders the cells and splits those that lie across the property boundary, while every other unknown cell is split along the state dimension that most changes the controller's actions along the simulated closed loop, i.e., the one that most reduces the pixel uncertainty reaching the controller. On a CARLA braking system and three Gym benchmarks, the DWM is more accurate to the real camera than a cGAN surrogate and yields 1.9x to 69x narrower robustness intervals. The vision part causes most of the remaining overapproximation, and our splitting reduces the unknown fraction of the four-dimensional CartPole initial set from 21.88% to 8.18% within the budget.
Concept vectors aim to enhance model interpretability by linking learned internal representations with human-understandable semantics, but their practical utility is often limited by noisy and inconsistent activations. In this work, we discover the SuperActivator Mechanism: a transformer dynamic that amplifies concept activation gaps, concentrating the most reliable concept evidence into a small set of highly activated tokens, which we call SuperActivators. We theoretically characterize this mechanism, proving that concept-aligned attention produces two competing effects: a tail amplification effect, where gaps among already-high activations grow disproportionately fast, and an equalization effect that eventually dominates as attention concentrates on extreme tokens. We validate the predicted tail amplification empirically, finding that in-concept activation distributions develop well-separated upper tails consistently across models, image and text modalities, and concept types. Moreover, we demonstrate the practical implications of SuperActivators for concept detection and localization. First, using only high-activation tokens as concept indicators improves detection by up to 14 F1 points over competing baselines. Second, restricting detection to the extreme activation tail is not merely sufficient but essential: including activations beyond the top 5-10% of the in-concept distribution introduces noise that degrades performance. Finally, using SuperActivators as instance-specific attribution targets yields more accurate and faithful concept localization heatmaps than using global concept vectors.
Reinforcement learning (RL) faces a persistent tension: policies that are stable to optimize (e.g., Gaussians) are often too simple to represent the multimodal action distributions required for complex control. Conversely, expressive generative policies---such as diffusion and flow matching---can be difficult to optimize in online RL due to intractable likelihoods and gradients propagating through long sampling chains. We address this tension with a key structural principle: decoupling optimization from generation. Building on this, we introduce GoRL (Generative Online Reinforcement Learning), an algorithm-agnostic framework that trains expressive policies from scratch by confining policy optimization to a tractable latent space while delegating action synthesis to a conditional generative decoder. Using a two-timescale alternating schedule and anchoring decoder refinement to a fixed prior, GoRL enables stable optimization while continuously expanding expressiveness. Empirically, GoRL consistently outperforms unimodal and generative baselines across diverse continuous-control tasks. Notably, GoRL achieves returns exceeding 870 on HopperStand, more than $3\times$ the strongest baseline; on high-dimensional humanoid tasks, it further outperforms the strongest non-GoRL baseline by over an order of magnitude. Code is available at https://github.com/bennidict23/GoRL.
Algorithms have been estimated to increase AI pretraining FLOP efficiency by a factor of 22,000 between 2012 and 2023 (Ho et al., 2024). Running ablation experiments on key innovations from this time period, we are able to account for less than 10x of these gains at small compute scales. Surveying the broader literature, we estimate that additional innovations not included in our ablations also account for less than 10x, yielding less than 100x efficiency gains total in low-compute regimes. This leads us to conduct scaling experiments, which provide evidence that the efficiency gap can be explained by algorithms with scale-dependent efficiency improvements. In particular, we conduct scaling experiments between LSTMs and Transformers, finding evidence for exponent differences in their compute-optimal scaling law while finding little scaling difference for many other innovations. These experiments demonstrate that, contrary to standard assumptions, an algorithm's efficiency gains are tied to compute scale. Using experimental scaling extrapolations and literature estimates, we can account for the majority of efficiency gains over the same time period. Our results indicate that algorithmic progress at smaller training compute levels has been far slower than previously assumed, and that measures of algorithmic efficiency are strongly reference-dependent.
Understanding reflection remains a long-standing challenge in 3D reconstruction due to the entanglement of appearance and geometry under view-dependent reflections. In this work, we present the Pygmalion Effect in Vision, a novel framework that metaphorically "sculpts" reflective objects into clay-like forms through image-to-clay translation. We then introduce a dual-branch design in which a BRDF-based reflective branch and a clay-guided branch share the same Gaussian geometry but operate through independent rendering paths. The clay-guided branch, supervised by the synthesized clay images, provides reflection-free geometric guidance that complements the photometric supervision of the BRDF branch. Experiments on both synthetic and real datasets show consistent improvements in both geometry and photometric metrics. Beyond technical gains, our framework reveals that seeing by unshining, translating radiance into neutrality, can serve as a powerful inductive bias for reflective object geometry learning.
A 3D transcriptomic profile of a tissue can give a holistic view of its organization in health and disease, but spatial transcriptomics (ST) is costly, so usually only some serial sections are profiled, while H&E is available for all. Most methods that infer ST from histology treat each section independently and ignore 3D structure, while existing 3D-aware approaches are not generative and do not scale well. We present Holographic Tissue Expression Inpainting and Analysis (HoloTea), a 3D-aware flow-matching framework that imputes spot-level gene expression on an unprofiled section from its H&E and the measured expression of adjacent profiled sections. Our key idea is to take the measured expression at the same registered position on neighboring sections and feed this cross-section context to the flow model through a lightweight ControlNet, so that predictions follow anatomical continuity along z. To capture the discrete, zero-inflated nature of counts, the flow starts from a learned zero-inflated negative binomial (ZINB) prior that, on an unprofiled section, is predicted from the measured expression of the nearest profiled section. A global attention block with a small set of inducing tokens adds whole-slide context at a cost linear in the number of spots, so training and inference scale to large 3D ST datasets. Across three ST datasets spanning different tissue types and resolutions, HoloTea consistently improves 3D expression accuracy and generalization compared to 2D and 3D baselines. We envision HoloTea advancing the creation of accurate 3D virtual tissues, ultimately accelerating biomarker discovery and deepening our understanding of disease. Code is available at https://github.com/Lotfollahi-lab/HoloTea.
Learning-based underwater image restoration and enhancement rely heavily on high-quality paired training data. However, obtaining clear reference images for real underwater scenes remains challenging. Existing datasets often use the outputs of enhancement algorithms as references, which may introduce uncertainty into supervision targets. To address this issue, we propose a generative data framework that preserves clear source images as restoration references and synthesizes corresponding underwater inputs through unpaired image-to-image translation. Using representative underwater appearances to guide the translation, we construct UWNature from natural photographs. We systematically examine the reliability of the generated supervision through its agreement with real underwater appearances, structural consistency, and depth-related degradation characteristics. Experiments across 6 restoration architectures and 3 test sets evaluate its effectiveness through objective metrics, human perceptual assessment, and the correspondence between synthetic degradation and learned restoration behavior. The results demonstrate competitive restoration performance and favorable perceptual quality on real underwater images. Further analysis reveals that greater spatial plausibility of synthetic degradation does not necessarily improve restoration performance, highlighting the importance of evaluating both the intrinsic properties of synthetic data and their effectiveness for model learning. These findings support a scalable framework for constructing and evaluating reliable paired supervision for underwater image restoration and enhancement. The generated datasets are available at: https://github.com/tianyufengcn/SynUIEDatasets.
Foundation models for EEG analysis are still in their infancy, limited by two key challenges: (1) variability across datasets caused by differences in recording devices and electrode configurations, and (2) the low signal-to-noise ratio (SNR) of EEG, where neural activity is often buried under artifacts and non-brain sources. To address these challenges, we present EEG-X, a device-agnostic and noise-robust foundation model for EEG representation learning. EEG-X introduces a location-based channel embedding that enables robust transfer across heterogeneous EEG devices and electrode layouts. To improve robustness against noise, EEG-X employs a novel noise-aware masking-reconstruction strategy that reconstructs artifact-cleaned signals rather than raw noisy EEG and introduces a novel Dictionary-based Convolutional Transformation (DiCT) that improves reconstruction-based pretraining by comparing signals in a structured feature space instead of directly in the raw signal space. Experiments across datasets collected from diverse EEG devices show that EEG-X consistently outperforms state-of-the-art methods across multiple downstream tasks and demonstrates strong cross-domain generalization when pretraining and downstream datasets differ in electrode layouts and recording configurations paving the way towards foundation models. The models and code are available at https://github.com/Emotiv/EEG-X.
Image-guided zero-shot style transfer enables training-free stylization of arbitrary content images using a reference style. Despite recent progress, existing methods commonly rely on self-attention manipulation to inject style information, which may transfer semantic content from the style image, leading to semantic leakage. In this work, we revisit this phenomenon from the perspective of key--value coupling in the self-attention of diffusion models. Our analysis shows that the native correspondence between style keys and values plays an important role in transferring semantic information from the style image. Motivated by this observation, we propose SuRe, a timestep-aware key--value regulation framework that mitigates semantic leakage by disrupting the native key--value correspondence while progressively restoring it for faithful style transfer. Specifically, we present value shuffle to decouple style keys and values while preserving the empirical distribution of the values, together with a progressive recoupling mechanism that gradually recouples them throughout the stylization process. Experiments against 18 representative methods demonstrate that SuRe maintains competitive style similarity while consistently improving content similarity across SD1.5, SDXL, SD3, and FLUX.1-dev, with natural generalization to multi-style transfer. The code will be publicly available at https://xinr-tang.github.io/SuRe-homepage/.
Video generation models have advanced significantly through the latent diffusion transformers trained with rectified flow techniques. Yet these models still struggle with geometric inconsistencies, unstable motion, and visual artifacts that break the illusion of realistic 3D scenes. 3D-consistent video generation could significantly impact numerous downstream applications in generation and reconstruction tasks. We explore how epipolar geometry constraints improve modern video diffusion models. Despite using massive training data, these models fail to capture fundamental geometric principles. We align diffusion models using pairwise epipolar geometry constraints via preference-based optimization, directly addressing unstable trajectories and geometric artifacts through mathematically principled geometric enforcement. Our approach efficiently enforces geometric principles without requiring end-to-end differentiability. Evaluation demonstrates that classical geometric constraints provide more reliable preference signals than the video-quality and geometry metrics based on foundational models. We construct preferences from static scenes with camera motion, where the epipolar metric is most reliable, while the aligned model generalizes to dynamic scenes. By bridging data-driven learning with classical computer vision, we reduce epipolar error by 31% and improve human-rated consistency from 54% to 72% without compromising visual quality.
Deep hashing encodes high-dimensional image features into compact binary codes, enabling efficient similarity search in large-scale content-based image retrieval, a problem akin to semantic-preserving lossy compression under extreme bit-rate constraints. A fundamental challenge in deep hashing stems from the discrete nature of quantization in generating the codes. W-type regularizations, such as $z\mapsto\||z|-1\|_1$, are widely adopted and empirically effective at pushing network outputs toward binary values. However, existing methods often directly optimize these regularizations without convergence guarantees. While proximal gradient methods offer a promising solution, the coupling between W-type regularizers and neural network outputs results in composite forms that generally lack closed-form proximal solutions. In this paper, using Fenchel duality, we reformulate the nonconvex W-type regularization term through its convex conjugate, yielding a min-max problem with tractable proximal updates. We show that this min-max problem captures the stationarity of the original deep hashing objective up to the quantization error. Building on this problem, we present a stochastic primal-dual hashing algorithm, referred to as DualHash, with oracle complexity guarantees. We derive two algorithm instances: a momentum-based version with $O(\varepsilon^{-4})$ oracle complexity and an improved $O(\varepsilon^{-3})$ version using variance reduction, both for $\varepsilon$-KKT points of the min-max problem, which yield near-stationarity guarantees for the original objective. Experiments on three standard datasets demonstrate competitive retrieval accuracy and reduced quantization error, validating the practical value of DualHash for image retrieval problems. Our implementation of DualHash is available at https://github.com/RailLee/DualHash.
Causal modeling offers a principled foundation for uncovering stable, invariant relationships in time-series data, thereby improving robustness and generalization under distribution shifts. Yet its potential is underutilized in spatiotemporal Earth observation, where models often depend on purely correlational features that fail to transfer across heterogeneous domains. We propose RIC-TSC, a regionally-informed causal time-series classification framework that embeds lag-aware causal discovery directly into sequence modeling, enabling both predictive accuracy and scientific interpretability. Using multi-modal satellite and reanalysis data-including Sentinel-1 microwave backscatter, Sentinel-2 and Landsat-8 optical reflectance, and CARRA meteorological variables-we leverage Joint PCMCI+ (J-PCMCI+) to identify region-specific and invariant predictors of supraglacial lake evolution in Greenland. Causal graphs are estimated globally and per basin, with validated predictors and their time lags supplied to lightweight classifiers. On a balanced benchmark of 1000 manually labeled lakes from two contrasting melt seasons (2018-2019), causal models achieve up to 12.59% higher accuracy than correlation-based baselines under out-of-distribution evaluation. These results show that causal discovery is not only a means of feature selection but also a pathway to generalizable and mechanistically grounded models of dynamic Earth surface processes.
Generative AI relies on statistical learning and local gradient-based guidance, often resulting in artifacts that may be more emulative than genuinely creative. This position paper advances a perspective on Evolutionary Computation (EC) as a paradigm of Natural Generative AI (NGAI), in which stochastic variation, diversification, and natural selection present an alternative foundation for generative exploration and creative synthesis. EC provides an array of variation operators with varied generative capabilities: parent-centric operators induce population distributions analogous to conventional generative models, while disruptive operators can produce structured evolutionary leaps to out-of-distribution artifacts (e.g., programs, designs, or AI agents themselves). This lens reveals a unique theoretical association between certain disruptive variations in multitask EC and a product-of-distributions population structure. A mathematical result derived herein shows that a product structure also occurs under a local second-order approximation of an established generative adversarial program for creative extrapolation, inspired by Martindale's cognitive theory of artistic creation. The simultaneous existence of multiple fitness landscapes in multitask EC inherently moderates selection pressure, allowing these novel artifacts to survive and seed further innovation. Structured disruption and selection moderation are thus argued to be key elements to support creative synthesis. Agnostic to any specific algorithmic realization, the perspective positions NGAI as an emerging engine for open-ended scientific exploration.
Generative models have recently emerged as powerful surrogates for physical systems, demonstrating increased accuracy, stability, and/or statistical fidelity. However, most approaches rely on iteratively denoising a Gaussian, a choice that may not be the most effective for autoregressive prediction tasks, where the current and future states are often closely related. We therefore benchmark generative models based on stochastic interpolants, which can learn maps between arbitrary densities, against standard Gaussian-denoising frameworks across a range of fluid and climate systems. Following prior work, we consider a transport map for dynamical systems that learns an interpolant between a point mass at the current state and a conditional distribution of future states. We find that stochastic interpolants can match or outperform Gaussian-denoising baselines in pointwise accuracy, spectral fidelity, and probabilistic skill. Furthermore, we identify a tradeoff where deterministic sampling can improve pointwise accuracy but smooth predictions, whereas stochastic sampling recovers spectral content and calibration but requires more function evaluations, as well as ensemble averaging to recover pointwise accuracy. As a whole, these results establish stochastic interpolants as a competitive baseline for generative physical emulation, and also indicate that the training loss, sampler, and transport map should be chosen with the physical system and target metric in mind.
Activation-aware low-rank factorization techniques yield strong compression results, but they are generally confined to linear layers, and existing whitening-based theory typically makes an implicit full-rank assumption on layer activations. We introduce a layer representation framework that allows us to understand and extend low-rank factorization beyond linear layers, including standard and grouped convolutions. Within this framework, we provide a broader characterization of whitening-based activation-aware factorization that naturally covers rank-deficient activations, and show that the resulting projection is optimal in terms of layer output distortion. We then use output distortion as a proxy to allocate per-layer ranks under explicit FLOP or parameter count budgets, with negligible overhead via a Lagrangian relaxation of the allocation problem. Together, these components form BALF, an end-to-end, fine-tuning-free pipeline for model compression. Across CNNs and vision transformers on CIFAR-10 and ImageNet-1K, BALF generally achieves higher accuracy than SVD-based factorization baselines at matched FLOP or parameter count targets, and remains competitive with other fine-tuning-free compression techniques.
One of the largest drivers of social inequality is unequal access to personal tutoring, with wealthier individuals able to afford it, while the majority cannot. Affordable, effective AI tutors offer a scalable solution. We focus on adaptive learning, predicting whether a student will answer a question correctly, a key component of any effective tutoring system. Yet many platforms struggle to achieve high prediction accuracy, especially in data-sparse settings. To address this, we release the largest open dataset of professionally marked formal mathematics exam responses to date. We introduce a probabilistic modelling framework rooted in Item Response Theory (IRT) that achieves over 80 percent accuracy, setting a new benchmark for mathematics prediction accuracy of formal exam papers. Extending this, our collaborative filtering models incorporate topic-level skill profiles, but reveal a surprising and educationally significant finding, a single latent ability parameter alone is needed to achieve the maximum predictive accuracy. Our main contribution though is deriving and implementing a novel discrete variational inference framework, achieving our highest prediction accuracy in low-data settings and outperforming all classical IRT and matrix factorisation baselines.