New papers on Other machine learning
165 new papers on other machine learning in the last 7 days, within AI & machine learning. These are the 50 Pipette rates most worth reading, with the main result in the authors' own words.
The best of the week
A generalizable structural brain MRI foundation model built through dual-priority federated pretraining
Across 20 downstream datasets spanning 17 classification, regression and segmentation tasks, BrainFedFM achieved the state-of-the-art performance (mean rank 1.68, 50% gain) across seven models, including four centralized foundation models, while showing particularly consistent advantages in classification and regression and robustness across underrepresented populations.
PreprintReal-world useTimeInteract: Towards Real-Time Interactive Intelligence for Streaming Time Series
We introduce a new regime, Time-Series Interaction: a model continuously perceives incoming time-series observations and user intent, autonomously decides when to remain silent or respond, and continues processing new observations during response generation.
PreprintComplex-valued Phase-Coherent Transformers
The resulting family of phase-coherent Transformers (\PCT) matches or exceeds the strongest real-valued baseline across long-range memory, positional retrieval, hierarchical reasoning, frequency-domain classification and physical complex signals; it shows no degradation up to depth 20; and its loss decreases log-linearly over a 61-fold range of parameters.
PreprintBold claims, read criticallyClaims a big stepDynamics-informed machine learning for recovering extensive missing systems dynamics
In this study, we introduce a dynamics-informed imputation framework, termed Full-Partial Reconstruction Mapping (FPRM).
Peer-reviewed journalReal-world useTimeBraid: Unifying Time Series and Language for Understanding and Forecasting
We present TimeBraid, a series of unified time-series and language models that align pretrained language models and pretrained time-series foundation models through interleaved global residual attention layers.
PreprintAsk for Any Appliance: A Prompt-Programmable Foundation Model for Non-Intrusive Load Monitoring
We present FM4NILM (Foundation Model for NILM), a single prompt-programmable model that estimates a requested appliance's power trajectory from aggregate measurements, a natural-language description, and optional activation exemplars.
PreprintReal-world useBrain-Inspired Hierarchical Modularity for General Continual Learning
Across visual recognition, vision-language understanding, ego-exo video understanding, and embodied vision-language-action learning, our method consistently improves learning under online and uncertain data streams, with gains exceeding 50 percentage points over replay-free alternatives in embodied manipulation.
PreprintBold claims, read criticallyLearning Where to Look: A Shared Relative-Alignment Module for Time-Series Forecasting and PPG-to-Vital-Sign Reconstruction
We propose ROOSTER, one conditioning module that handles vital-sign reconstruction and time-series forecasting alike by learning this correspondence.
PreprintReal-world usePrior-Amortized In-Context Bayesian Inference for Generalized Linear Mixed-Effects Models
We introduce metabeta, a pretrained neural network for prior-amortized in-context Bayesian inference over GLMMs.
PreprintScalarLens: Numerical Embeddings with Stable Coordinates and Contextual Responses for CTR Prediction
We introduce ScalarLens, a numerical embedding that preserves what a value is while adapting how it should be interpreted.
PreprintSigned Graph Pre-Training and Prompt Learning
In this paper, we introduce TopoSIGN, a pioneer topology-guided graph pre-training and prompt learning framework for signed graphs.
PreprintMIND the Gap: A Geographic Implicit Neural Representation with Adjustable Spatial Scale
We introduce Matryoshka Implicit Neural Distillation (MIND), which distills embeddings from specialist pretrained geospatial models into a single generalist coordinate embedding with adjustable spatial granularity.
PreprintDiaSeg: Diagonal Segment Extraction from DTW Paths for Interpretable Gait Analysis
We introduce DiaSeg, a framework that extracts diagonal segments from DTW paths with controlled breaks, characterizing each segment by five geometric features (effective length, interruption count, cost variation, temporal position, and path context), and enabling unsupervised pattern discovery without domain-specific feature engineering.
PreprintReal-world useBrain-Token Learning: Microstate-Based Tokenization and Multi-Scale Interaction for Long-Horizon EEG Sequence Modeling
In this work, we propose Brain-Token Learning, a neuroscience-inspired framework that introduces Brain Tokenization for long-horizon EEG sequence modeling.
PreprintGrowth-Inspired Graph Generation and Inverse Design of Mechanical Lattices via Dot Matrices Database Augmentation and GCNN
Inspired by this developmental logic, this work introduces a morphogenetic graph-generation framework for mechanical lattices in which a discrete dot matrix provides potential nodes and the final architecture is created by sequential cross-layer and intra-layer growth.
PreprintBold claims, read criticallyExponential Family Synthetic Controls
We develop exponential family synthetic controls (EFSC), a distributional version of synthetic controls for a panel of datasets.
PreprintReal-world useCode availableITSY: Causal Discovery From Irregular Time-Series Data
We introduce ITSY, the first continuous-optimization method for causal discovery from irregular time series under a linear model.
PreprintAirGC-CD: Gaussian-Circulant Precoding for Exactly Debiasable PAPR Reduction in Over-the-Air Federated Learning
To address these challenges, we propose AirGC-CD, an over-the-air scheme that precodes each local update with a partial Gaussian circulant matrix before clipping.
PreprintReal-world useDefaultGNN: A Dual-Perspective GNN Framework for Predicting Corporate Default from Buyer-Seller Transaction Networks
DefaultGNN integrates both views to model how risk flows through transactional relationships, achieving strong improvements over both attribute-based and graph-based baselines, especially for firms with limited intrinsic risk signals.
Preprint with a published versionCode availableActionable Insights from Observational Data: The Case of Advanced Classes in K-12 Education
Our analysis shows that enrolling in advanced English courses has a net positive but modest effect on student achievement outcomes.
PreprintCOMPLEX: A Closed-Form Certified Embedding of Multiparameter Persistence Modules
COMPLEX is a closed-form, training-free embedding of multiparameter modules -- slice the module along a fixed near-diagonal net, embed each slice barcode by the certified PLACE/PALACE landmark map, concatenate.
PreprintBold claims, read criticallyClaims a big stepEvent Signature Transfer: Model-Agnostic Forecast Scenario Construction from Historical Events
We introduce Event Signature Transfer (EST), a training-free, model-agnostic operator that turns a completed past event into an explicit forecast scenario.
PreprintDiscovery-Driven Integration of Disjoint Tables via Text
Comprehensive evaluations on real-world benchmarks demonstrate that LOKI consistently outperforms state-of-the-art multi-modal data discovery approaches, and materializes typed integrated tables with 0.982 macro typed-pair precision while being up to 40 times cheaper in LLM API cost than direct prompting.
PreprintCTRL: Control-Based Time Series Forecasting with LLM-Guided Residual Learning
We introduce CTRL, a framework that decouples semantic reasoning from quantitative prediction.
Preprint with a published versionhyperbolix: Hyperbolic Deep Learning in JAX
We present hyperbolix, an open-source library for hyperbolic deep learning in JAX, built on Flax NNX.
PreprintCode availableConditional Tensor Diffusion: Distributional Counterfactual Learning and Inference
We develop Counterfactual Tucker Diffusion (\CFTDiff), which integrates the treatment mask and latent Tucker structure into conditional diffusion to recover this distribution given observed control outcomes through efficient nonlinear score learning in a low-dimensional core.
PreprintDecoupled Causal Discovery
While existing approaches are primarily based on conditional independence tests, structure scores, or restrictive functional assumptions, we propose Decoupled Causal Discovery (DCD), a novel decoupling-based perspective that does not rely on these methodologies.
PreprintFed-ReMasker: Federated Tabular Imputation under Feature-Level Missingness
To address this setting, we adapt the ReMasker masked autoencoder to federated learning (Fed-ReMasker), enabling centers to impute features never observed locally by leveraging knowledge learned across collaborating centers.
PreprintReal-world useWhen Tomorrow Becomes Today: Self-Evolving Policies for Agentic Time-Series Forecasting
To exploit this delayed feedback systematically, we introduce TimEvolve, a frozen-backbone time series agent that converts each realized outcome into persistent joint updates of expert trust, agent path selection, and intervention strength.
PreprintFedIncome: Federated Learning for Income Estimation in Digital Lending Under Data Sovereignty Constraints
We introduce FedIncome, a federated learning framework for income estimation that enables institutions to train a shared model without pooling raw borrower records.
PreprintReal-world useShapeLex: Decoupling Local Shape Symbolization and Global Scale Modeling for Text-Controlled Time Series Generation
We propose Shape Lexicon (ShapeLex), which decouples text-to-sequence generation into discrete symbolization of local shapes and continuous modeling of global attributes.
PreprintWhat Do Tabular Foundation Models Compute In Context? In-Situ Representation Refinement through Attention-Gated Updates
We develop in-situ representation refinement: support labels guide updates to the episode's representations, and these updates transfer to unlabeled queries without changing model parameters.
PreprintMerge++: Universal Merge Refinement Through Data-Free Checkpoint Inversion
We propose Merge++, a post-hoc method that addresses this by inverting the expert checkpoints to synthesize task-representative images, then distilling expert knowledge into the merged model using those images.
PreprintBold claims, read criticallyICE: Task-Aligned Clifford Latent Fields for Multimodal Graph Foundation Models
We therefore introduce ICE (Interaction-aware Clifford Encoder), a multimodal graph foundation model built on a node-indexed Clifford latent field.
PreprintBold claims, read criticallySGA: Uncertainty Quantification for Multi-Step Forecasting in Time Series Foundation Models
In this paper, we propose the Slicing-Graphing-Alignment (SGA) method to quantify the uncertainty of multi-step TSFM forecasts.
PreprintFrom Graphs to Feeders: Constraint-Guided Diffusion for Rule-Compliant Feeder Generation
We therefore formulate feeder synthesis as a constraint-guided graph generation problem and propose the Power-Grid-constrained Discrete Denoising Diffusion model, PG-DiGress, which learns categorical node and edge patterns from feeder data, while respecting domain-specific rules.
PreprintReal-world useBenchmarking Hybrid Deep Learning Architectures for Predictive Maintenance in Industry 4.0
We also found that the hybrid method that combines a Long Short-Term Memory (LSTM) layer with a Transformer layer is more resilient to noisy data from factory shops [7].
Preprint with a published versionReal-world useNeuralized Multi-Wavelet Decomposition for Time Series Classification and Forecasting
To address this gap, we propose m-WCN, a novel end-to-end deep learning framework that neuralizes multi-wavelet decomposition for joint extraction of temporal patterns and frequency components.
PreprintAugmented Hypothesis Testing with Persona-Based LLM Simulations
We propose a principled framework for learning-augmented hypothesis testing that leverages predictions of unknown quality to reduce sample sizes while maintaining statistical validity.
PreprintTNLearn: An Open Source Python Package for Task-based Neurons
To facilitate the use of task-based neurons in scientific research and industrial applications, we introduce TNLearn, an open-source Python package that provides automated construction of task-based neurons and networks, enabling smooth training of task-based networks.
PreprintCode availableBiView-Touch: Learning Bimanual Tactile Representations by Cross-Hand Completion
To exploit this overlooked structure, we introduce BiView-Touch, a tactile-only framework that completes masked target-hand latents from the remaining visible target-hand regions and the synchronized full contralateral hand.
PreprintBeyond Static Graph World Models: Learning Stochastic Latent Dynamics over Evolving Topologies
We propose the Graph Dynamics Model (GDM), a world model for graph-structured observations that is designed to handle the more general setting of evolving topologies in stochastic and partially observable environments.
PreprintEMGBlend: Heterogeneity-Aware Self-Supervised Pretraining for Gesture and Force Decoding
We introduce EMGBlend, a self-supervised framework designed around these differences.
PreprintReal-world useCode availableReproducible AI Requires Reproducible Randomness
Our results demonstrate that reproducibility cannot be assumed from PRNG state transfer alone, even when implementations claim to follow the same underlying algorithm.
PreprintReal-world useTopological Signal Processing With Unoriented Operators
We study an unoriented TSP (UTSP) framework that replaces oriented boundaries with unoriented incidence matrices.
PreprintGeoBalance: Geometry-Aware Monitoring and Reconstruction with Asymmetric Optimization for Balanced Multimodal Learning
Motivated by this observation, we propose GeoBalance, a geometry-aware framework that monitors these two geometric properties and reconstructs the weak modality representation only when it exhibits signs of MMC.
PreprintSpatio-temporally complementary feature propagation on graphs for longitudinal AADT estimation
This research proposes a novel spatio-temporally complementary feature propagation framework that leverages the strengths of two distinct data sources: spatially sparse but temporally dense loop detector data, and a spatially complete but temporally sparse macroscopic transportation model.
PreprintReal-world useProbabilistic Forecasting of Business Process Executions with Neural Temporal Point Processes
On ten public logs, the resulting model matches discriminative baselines on point accuracy, dominates them on the calibration and sharpness of remaining-time distributions, and is the cheapest at inference, since a full predictive distribution is obtained in a single forward pass without sampling.
PreprintAbility-Residual Decoupled Modeling for Affective Cognitive Diagnosis
To address this problem, we propose an ability-residual decoupled framework for affective cognitive diagnosis.
PreprintReal-world useLet Training Guide Selection: Online Synthetic Data Filtering via Real-Anchored Utility
We propose FROST, an online framework that estimates synthetic-data utility through gradient feedback anchored in real training data.
PreprintReal-world use