cs.LG · 2026-09-16 · No. 115

Machine Learning, 2026-09-16.

54 new papers in cs.LG. Titles, authors, abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →

01 — The papers

54 entries
  1. 01

    ENCP: Episode-Normalized Conformal Prediction for Vision-and-Language Navigation

    Vicky Feliren, A. Taufiq Asyhari, Muhamad Risqi U. Saputra

    cs.LG · cs.AI · cs.RO

    Uncertainty estimation for Vision-Language-Navigation (VLN) models is a critical task since it can help identify ambiguous and unreliable predictions, enabling agents to make safer navigation decisions. As one of the most advanced uncertainty estimation frameworks, conformal prediction (CP) offers a promising approach for uncertainty estimation in VLN. However, given that VLN agent requires a sequence of steps, standard calibration in...

    arxiv.org/abs/2609.17499 · PDF

  2. 02

    FreqSpaNet: Frequency and Spatial Learning of SFPF for Physical Layer Hardware Integrity Detection

    Xiaoxuan Huang, Jinlong Xu, YiZhe Wang, Meng Zhang, Xian Li, Yuying Bian

    cs.LG

    Unauthorized hardware replacement can preserve a wireless device's logical identity while altering its physical implementation, posing a challenge to hardware integrity verification. Spatio-frequency polarization fingerprints (SFPFs) capture device-dependent responses across multiple frequencies and directions, but their frequency and spatial dimensions exhibit different structural dependencies. We propose FreqSpaNet, an SFPF representation...

    arxiv.org/abs/2609.17491 · PDF

  3. 03

    Coupled Calibration and Learning: Mitigating Teacher Bias in LLM Distillation without Target-Domain Reward Feedback

    Haichen Hu, Yuheng Zhang, David Simchi-Levi

    cs.LG · cs.AI · math.ST · stat.ML

    Large language model (LLM) distillation aims to transfer the capabilities of a powerful teacher to a smaller student. Direct imitation, however, can also transfer the teacher's systematic bias and errors. This challenge is particularly pronounced under covariate shift, when the teacher's reliability on target questions is uncertain and target-domain reward feedback is unavailable. We propose Coupled Calibration and Learning (CCL), an LLM...

    arxiv.org/abs/2609.17474 · PDF

  4. 04

    Reduced-Space Multi-Fidelity Bayesian Optimization of Process Simulation Models

    Niki Triantafyllou, Andrea Bernardi, Maria M. Papathanasiou

    cs.LG · math.OC

    Optimizing industrial process flowsheets is often computationally prohibitive due to the high cost of rigorous simulations and the curse of dimensionality inherent in complex design spaces. To address these challenges, we present a reduced-space multi-fidelity Bayesian optimization (RS-MFBO) framework designed for high-dimensional, expensive black-box functions. The approach integrates Global Sensitivity Analysis (GSA) for dimensionality...

    arxiv.org/abs/2609.17440 · PDF

  5. 05

    Learning-Guided Planning in Large Dynamic Action Spaces: Budgeted Tree Search for One-to-Many Mobile Charging

    Liang-Ching Tao, Pi-Chung Wang

    cs.LG · cs.AI

    Many learned sequential decision systems map the current state directly to an action. That shortcut becomes brittle when candidate actions are numerous, geometrically structured, and rebuilt with the state. One-to-many mobile charging makes this setting concrete: with N=250 sensors, the initial state induces about 1,125 candidate charging-stop actions; each chosen stop simultaneously serves its in-range sensors, and the action universe...

    arxiv.org/abs/2609.17429 · PDF

  6. 06

    Bridging the Confidence Gap: Temperature Scaling for Calibrating Test-Time Prompt Tuning

    Yuwei Liang, Jian Liang, Dapeng Hu, Yinuo Xu, Ran He

    cs.LG

    Test-time prompt tuning (TPT) enables adaptation on a single test instance, achieving improved accuracy but often sacrificing calibration performance. Most existing calibration methods introduce additional regularization terms to promote dispersion across text embeddings and reduce calibration error, yet these methods often suffer from a drop in accuracy. Motivated by the well-calibrated nature of zero-shot predictions, we propose CoTS, a...

    arxiv.org/abs/2609.17386 · PDF

  7. 07

    OPEN-1B: A Fully Auditable Training Run

    John Donaghy, Brian Wilcox, Oğuzhan Ersoy, Shikhar Rastogi, Adam St Arnaud, Alexey Titov, Jordan Greenberg, Ben...

    cs.LG

    Open-source language models have a reproducibility problem. Despite releasing weights, training data, and recipes, none of them are provably reproducible due to the non-associativity of floating-point arithmetic. Deep learning frameworks often offer a deterministic execution mode, allowing reproducible operations on the same machines. Unfortunately, this determinism does not carry across hardware such that a user can verify that a released...

    arxiv.org/abs/2609.17380 · PDF

  8. 08

    Large Language Models Develop Belief State Geometry In-Context

    Daniel Balcells, Andrew Jun Lee, Chirag Rastogi, Paul M. Riechers, Adam Shai, Xavier Poncini

    cs.LG · cs.CL

    Large language models (LLMs) trained on next-token prediction exhibit remarkable in-context learning (ICL) abilities, yet the representations that support ICL remain poorly understood. We consider such representations in a controlled setting: prompting LLMs with data emitted from hidden Markov models (HMMs) and probing for the corresponding belief state -- the posterior distribution over the HMM's hidden states given the observed token...

    arxiv.org/abs/2609.17376 · PDF

  9. 09

    Hybrid Variational Quantum Circuits for Multivariate Regression and High-Dimensional Data Reconstruction

    Koffi Ognandon Ayena, Frédéric Holweck, Serge Iovleff, Amah S d'Almeida

    cs.LG · stat.ML

    Variational quantum circuits (VQCs) are parameterized quantum circuits optimized classically. We propose a hybrid variational quantum circuit (HVQC) extending VQCs with a classical affine post-measurement layer, enabling vector-valued regression without the linear overhead of independent scalar circuits. Theoretically, we show that elementary one-and two-qubit circuits can approximate quadratic functions and products via data re-uploading and...

    arxiv.org/abs/2609.17358 · PDF

  10. 10

    Same Flow, Different Paths: Variance Reduction in Flow Matching

    Alexander Tyurin

    cs.LG · math.OC

    In flow matching (FM), a velocity model $v_θ$ is trained using a predefined path $g_t$ that connects data and noise samples (e.g., $g_t(x_0, x_1) = (1 - t) x_0 + t x_1$). In this work, we study the choice of this path from an optimization perspective by analyzing the variance of stochastic gradients. We consider the class $G(p_t,v^\star_t)$ of paths that induce the same marginal distributions $p_t$ and marginal velocity field $v^\star_t$, and...

    arxiv.org/abs/2609.17287 · PDF

  11. 11

    Personalized Federated Learning through Global Knowledge Distillation and Local Head Adaptation

    Polycarpo Souza Neto, José Mairton Barros da Silva Júnior, Charles Casimiro Cavalcante

    cs.LG · stat.ML · stat.OT

    Statistical heterogeneity limits federated learning when a single global classifier cannot represent client-specific label distributions. In this work, we propose Personalized Federated Knowledge Distillation with Head Adaptation (pFedKDH), which aggregates only the shared backbone, keeps persistent client-specific heads, and uses a recalibrated global head as a teacher during local training. Across MNIST, Fashion-MNIST, CIFAR10, and CIFAR100...

    arxiv.org/abs/2609.17284 · PDF

  12. 12

    Easy to Catch a Liar, Hard to Clear an Honest One: Language Models Diagnosing a Corrupted Reward Channel from a Verified Record

    Arman Nik Khah

    cs.LG · cs.AI · cs.CL

    An agent that learns from rewards has to trust whatever reports those rewards. When the reports suddenly change, either the world changed or the reporter broke. From the reports alone these are indistinguishable, and reinforcement learning theory shows that no amount of further experience separates them. The prescribed escape is richer data about the reporter itself. We ask whether a frozen language model, handed exactly that data, uses it....

    arxiv.org/abs/2609.17226 · PDF

  13. 13

    Memorisation bias in medical AI

    Moritz A. Knolle, Martin J. Menten, Laurin Lux, Mélanie Roschewitz, Emma A. M. Stanley, Georgios Kaissis, Daniel...

    cs.LG · cs.CY

    Medical AI models hold immense potential to improve patient outcomes, but they are also known to unintentionally memorise individual records from their training datasets. While such memorisation has been linked to targeted privacy attacks, its consequences for clinical deployment, where patients may be assessed by a model that saw their historical data during training, remain poorly understood. Here we show that predictions on a patient's...

    arxiv.org/abs/2609.17223 · PDF

  14. 14

    MyoFlow: Anchor-Tied Rectified Flow for HD-sEMG Gesture Recognition Across Sessions and Subjects

    Chenhao Wu, Dingjie Peng, Satoshi Funabashi, Satoshi Konishi, Wuqiang Yang, Hiroshi Onoda, Hironori Washizaki, Jiang Liu

    cs.LG

    High-density surface electromyography (HD-sEMG) gesture recognition supports prosthetic control, assistive robotics, and rehabilitation, but electrode re-donning and physiological variability cause distribution shifts that degrade accuracy across sessions and subjects. Generative HD-sEMG models primarily synthesize signals for augmentation; although diffusion models enhance representation learning, prediction still relies on a separate...

    arxiv.org/abs/2609.17194 · PDF

  15. 15

    LoopSpec: Pipelined Self-Speculative Decoding for Looped Transformers

    SangLyul Cho, Langqing Cui, Sehoon Kim, Dongsu Han, Insu Han

    cs.LG · cs.CL

    Looped Transformers achieve strong performance with compact parameter sizes by repeatedly applying a shared stack of Transformer blocks across recurrent depths. However, they incur higher decoding latency than standard Transformer models of comparable parameter size because shared weights are accessed at every recurrent depth. To improve decoding efficiency, self-speculative decoding is particularly well suited to Looped Transformers, as...

    arxiv.org/abs/2609.17184 · PDF

  16. 16

    IRENE: A Convolutional GRU Ensemble Model for Radar Precipitation Nowcasting over Italy

    Alessandro Camilletti, Gabriele Franch, Elena Tomasi, Marco Cristoforetti

    cs.LG · physics.ao-ph

    We present IRENE (Italian Radar Ensemble Nowcasting Experiment), a deep learning model for probabilistic short-range precipitation nowcasting over the Italian domain at \SI{1}{km} spatial and 5 min temporal resolution. IRENE adopts an encoder--forecaster architecture built on multi-scale Convolutional Gated Recurrent Units (ConvGRUs), trained on the national radar composite produced by the Italian Civil Protection Department (DPC). An...

    arxiv.org/abs/2609.17175 · PDF

  17. 17

    A unified framework for global and local interpretability using adaptive derivative-ordered random explanation

    Lemen Chao, Ming Lei, Anran Fanga

    cs.LG · cs.AI

    The interpretability of complex machine learning models is of paramount importance, especially in real-world high-stakes domains such as healthcare and finance. However, existing post-hoc interpretability methods suffer from inherent limitations: fragmented analytical processes, inadequate capacity to model nonlinear feature interactions, computational inefficiencies, and over-reliance on specific model architectures. To address these...

    arxiv.org/abs/2609.17171 · PDF

  18. 18

    Neural Field Ensembles for Aerodynamic Surface Prediction: Winning Solution to the ONERA CRM Wall Distribution 2025 Challenge

    Lionel Salesses, Caroline Sainvitu, Tariq Benamara

    cs.LG · physics.flu-dyn

    Machine-learning surrogate models offer a promising alternative to high-fidelity Computational Fluid Dynamics (CFD) simulations for aerodynamic analysis and design. However, constructing accurate surrogates for realistic aircraft configurations remain challenging due to complex geometries, multiple flow regimes, and limited training data. This work presents the methodology that achieved first place in the ONERA CRM Wall Distribution...

    arxiv.org/abs/2609.17160 · PDF

  19. 19

    High-Fidelity Digital Twin Data Models by Randomized Dynamic Mode Decomposition and Deep Learning with Applications in Fluid Dynamics

    Diana A. Bistrian

    cs.LG · math.NA

    The purpose of this paper is the identification of high-fidelity digital twin data models from numerical code outputs by non-intrusive techniques (i.e., not requiring Galerkin projection of the governing equations onto the reduced modes basis). In this paper the author defines the concept of the digital twin data model (DTM) as a model of reduced complexity that has the main feature of mirroring the original process behavior. The significant...

    arxiv.org/abs/2609.17101 · PDF

  20. 20

    Repurposing Unified Topological Signatures for Graph Representation Learning

    Sanyam Sanjay Jain, Anshika Krishnatray, Aditya Sharma, Vinti Agarwal

    cs.LG · cs.AI

    Message-passing Graph Neural Networks (GNNs) iteratively propagate and aggregate local neighborhood information followed by global readout to learn graph representations. However, their discriminative power is upper-bounded by the Weisfeiler--Lehman (1-WL) graph isomorphism test. This prevents GNNs from distinguishing certain non-isomorphic graphs with identical local neighborhood structures, often leading to similar graph representations....

    arxiv.org/abs/2609.17061 · PDF

  21. 21

    Learning Options for Compositional Motor Control with Adapter Banks

    Sreejan Kumar, Marcelo Mattar, Lea Duncker

    cs.LG · cs.RO · q-bio.NC

    Learning flexible motor primitives is a hallmark of skilled motor control. Recent neuroscience theory proposes that motor primitives may be implemented as low-rank perturbations of a shared recurrent network, but leaves open how such a system is learned. We translate this principle into a novel architecture for learning motor skills end-to-end: a shared recurrent core modulated by a bank of residual adapters, each selected by a discrete...

    arxiv.org/abs/2609.17042 · PDF

  22. 22

    Distributed JEPA: A Self-Supervised Framework for Energy Forecasting

    Liana Toderean, Tudor Cioara, Vasilis Michalakopoulos, Efstathios Sarantinopoulos, Ionut Anghel, Elissaios Sarmas

    cs.LG · cs.AI

    Traditional energy forecasting solutions rely on task-specific supervision and energy asset representations, limiting transferability and the ability to capture general temporal dynamics across heterogeneous assets. We address this by proposing a distributed Joint Embedding Predictive Architecture (JEPA) for self-supervised learning from heterogeneous energy time-series. The framework predicts latent representations of masked temporal...

    arxiv.org/abs/2609.17029 · PDF

  23. 23

    CLARE: Scalable Class-Incremental Continual Learning via a Sparsity-Based Framework

    Yunxiang Fu, Meng Lou, Zicheng Liao, Yizhou Yu

    cs.LG · cs.CV

    Continual learning must balance the learning of new knowledge with the retention of previously learned knowledge to incrementally learn tasks from a data stream without catastrophic forgetting. While leveraging pretrained models has significantly advanced continual learning, existing methods exhibit a scalability bottleneck when trained sequentially on many tasks, suffering from performance degradation due to inter-task interference and loss...

    arxiv.org/abs/2609.17026 · PDF

  24. 24

    Structural Negative Transfer in Federated Graph Neural Networks: Diagnosis, Causal Investigation, and the Limits of Divergence-Aware Mitigation

    Chethana Prasad Kabgere, Shylaja SS

    cs.LG · cs.DC

    Federated learning lets multiple participants train a shared model without pooling raw data, by exchanging locally trained model updates instead. Federated averaging assumes that averaging local models is a reasonable way to solve one shared problem when participants' data are broadly similar. Work on non-IID federated learning has shown that this assumption can withstand differences in label and feature distributions. We ask whether it...

    arxiv.org/abs/2609.16977 · PDF

  25. 25

    Beyond Token-Local Imitation: Reward-Compatible Temporal Credit Assignment for On-Policy Distillation

    Shiqi Liu, Zeyu He, Letian Tao, Guojian Zhan, Jiaxin Gao, Feihong Zhang, Jingliang Duan, Wei Xiong, Kehua Sheng, Bo...

    cs.LG · cs.AI · cs.PL

    On-policy distillation (OPD) has emerged as an effective approach for large language model post-training, yet existing objectives face a trade-off between objective fidelity and optimization stability. Token-level OPD provides stable but local supervision, whereas sequence-level OPD captures future credit at the cost of horizon-dependent variance. We establish a unified temporal-credit view of these formulations, showing that practical...

    arxiv.org/abs/2609.16937 · PDF

  26. 26

    When Confidence Signals Disagree: Local and Global Confidence in Autoregressive Language Models

    Julio C. Amador Diaz Lopez

    cs.LG · cs.AI

    Modern predictive systems expose multiple quantities that are commonly interpreted as measures of confidence. However, these quantities can summarize different aspects of the predictive process. This distinction matters when confidence is used to evaluate reliability or inform downstream oversight and control. We investigate whether different confidence readouts are empirically interchangeable in an autoregressive language model by comparing...

    arxiv.org/abs/2609.16933 · PDF

  27. 27

    Repurposing Deep Limit Order Book Forecasting for Scenario-Conditioned Market Impact Modeling

    Eljas Linna, Kestutis Baltakys, Derrick Manoharan, Alexandros Iosifidis, Juho Kanniainen

    cs.LG · cs.AI

    Deep Limit Order Book forecasting models capture nonlinear market dynamics, but their ability to quantify the effects of counterfactual order book messages has not been systematically validated. We introduce a model-agnostic framework that compares a trained forecaster's predictive distributions before and after injecting mechanically valid counterfactual messages, defining short-horizon model-implied market impact. A Transformer-based...

    arxiv.org/abs/2609.16930 · PDF

  28. 28

    Verbalizing Subliminal Learning Effects Using Text Optimization

    Nathan Hu, Sanmi Koyejo, Christopher Potts

    cs.LG · cs.AI · cs.CL

    Subliminal learning is a phenomenon in which a distillation dataset transmits traits from the teacher model that are not legibly encoded in the dataset itself. This introduces a new challenge for model development and creates new risks from data poisoning. In this work, we use text optimization to detect subliminal learning effects and describe them as legible prompts. Subliminal learning from a prompted teacher motivates our approach. We...

    arxiv.org/abs/2609.16927 · PDF

  29. 29

    HyCoSeq: Contextual Hyperbolic Representation Learning for Genomic Sequences

    Chenhao Zeng, Zhibin Pu, Shufei Ge

    cs.LG · q-bio.GN · stat.ML

    Hyperbolic geometry provides a natural inductive bias for genomic representation learning, but existing hyperbolic genomic models primarily use Lorentz convolutions to learn local sequence representations, while their residual pathways do not directly aggregate full Lorentz representations. We propose HyCoSeq, a contextual hyperbolic representation learning framework for genomic sequences. HyCoSeq incorporates weighted Lorentzian residual...

    arxiv.org/abs/2609.16925 · PDF

  30. 30

    Can Deep Learning Achieve Cross-Physics Mapping?

    Pengfei Zhu, Julien Lecompagnon, Mathias Ziegler

    cs.LG · physics.app-ph

    Can deep learning translate physical fields governed by fundamentally different equations? We address this question by introducing Cross-Physics Mapping (CPM), an operator-learning framework for mappings between heterogeneous physical domains. We formulate sufficient conditions for such mappings through compatible latent representations and propose a dimensionless scaling principle that aligns the characteristic evolution scales of the source...

    arxiv.org/abs/2609.16853 · PDF

  31. 31

    Information Geometric Self-Organization at the Edge of Stability in High-Capacity Kernel Associative Memories

    Akira Tamamori

    cs.LG · cs.NE

    High-capacity associative memories based on Kernel Logistic Regression (KLR) exhibit exceptional storage capabilities and robustness. Previous empirical studies identified a hyperparameter regime, the "Ridge of Optimization," where attractor stability is maximized. However, the geometric nature of this regime and the optimization dynamics required to reach it have remained unclear. In this paper, we investigate the static geometry of the...

    arxiv.org/abs/2609.16827 · PDF

  32. 32

    Adapting to Decision-Relevant Non-Stationarity in Decentralized Heterogeneous Bandits

    Zhaojun Peng

    cs.LG

    Decentralized bandit systems often contain heterogeneous agents: rewards can change at individual agents even when the best action for the network stays the same. These local changes may cancel when rewards are averaged across agents, so the number of local changes $\Stloc$ can be much larger than the number of changes in the best common arm $\Stdec$. We introduce Decision-Relevant Fresh Comparison (DRFC), which uses new, balanced samples...

    arxiv.org/abs/2609.16824 · PDF

  33. 33

    LCAP: Population-Informed Latent Chip Adaptation from Few Output Probes for Photonic Neural Networks

    Tianyu Gao, Guantian Zheng

    cs.LG

    Photonic neural networks (PNNs) offer efficient analog inference, but parameters optimized under ideal device models can degrade after fabrication, creating a persistent simulation-to-hardware (sim-to-real) gap. When many identically designed chips are deployed, calibrating each device from scratch compounds this cost. We propose Latent Chip Adaptation from Probes (LCAP), a population-informed framework that decomposes hardware adaptation...

    arxiv.org/abs/2609.16823 · PDF

  34. 34

    ImpossibleRubrics: Stress-Testing Generated Rubrics as Reward Signals

    Bowen Qin, Yi Xie, Yesheng Liu, Xi Yang

    cs.LG · cs.CL

    Language model-generated rubrics are increasingly used as reward signals for rubric-based reinforcement learning, LLM-as-a-judge evaluation, and automated grading. Such rubrics are reliable only if they reward honest answers over adversarial answers optimized to exploit them. Yet their robustness to such optimization remains poorly understood. We isolate the hardest regime: impossible tasks, where the prompt pressures the model toward an...

    arxiv.org/abs/2609.16816 · PDF

  35. 35

    Geometry of learning dynamics: Gradient descent versus natural gradient on the ridge of optimization

    Akira Tamamori

    cs.LG · cs.NE

    High-capacity associative memories based on Kernel Logistic Regression (KLR) exhibit a "Ridge of Optimization" characterized by extreme stability and a highly skewed weight spectrum. However, the dynamical process by which learning converges to this critical regime has remained unclear. This paper provides a geometric analysis of the learning trajectories on the statistical manifold of a KLR-trained Hopfield network. By comparing the paths of...

    arxiv.org/abs/2609.16805 · PDF

  36. 36

    SOTER: A Generative Time-Series Foundation Model for Wearable Human Physiological Signals

    Fangke Chen, Sirry Chen, Wei Chen, Zhongyu Wei

    cs.LG · cs.AI

    Time-series foundation models have demonstrated strong cross-domain transfer, yet their common architectural assumptions remain poorly aligned with wearable physiological signals, which are multichannel, irregularly sampled, noisy, and governed by coupled continuous-time dynamics spanning distinct spectral scales. We present SOTER, a generative foundation model for wearable physiological time series that unifies cross-channel coupling,...

    arxiv.org/abs/2609.16804 · PDF

  37. 37

    Noise2Noise Revisited: Training Pair Distributions Dominate Loss Choice in Self-Supervised Denoising

    Dingyan Shang, Zhenyu Xu, Youting Wang, Bonan Shen, Bowen Liu

    cs.LG · cs.CV · eess.IV

    Noise2Noise (N2N) trains denoisers on pairs of independently corrupted observations, eliminating clean references. We stress-test two natural conjectures about why the L1 loss outperforms L2 here. First, the hypothesis that the L1 loss confers robustness via parameter sparsity confuses the loss with Lasso regularization: an explicit Lasso penalty produces the predicted sparsity yet fails to reproduce L1's cross-noise behavior, while L1- and...

    arxiv.org/abs/2609.16788 · PDF

  38. 38

    TAME: Token Attribution and Masking for Emergent misalignment

    Md Rayhanul Masud, Md Rizwan Parvez

    cs.LG · cs.AI · cs.CL

    Fine-tuning an aligned language model on narrow, flawed data can induce harmful behavior far outside the training domain, known as emergent misalignment (EM). Prior work has localized EM in model weights, activations, and training documents, but it remains unclear which training tokens carry the relevant fine-tuning signal. We introduce TAME (Token Attribution and Masking for Emergent Misalignment), a three-stage framework: token attribution...

    arxiv.org/abs/2609.16754 · PDF

  39. 39

    A Systematic Evaluation of Machine Learning Methods for Fault Detection and Line Identification in Electrical Power Grids

    Julian Oelhaf, Georg Kordowich, Paula Andrea Pérez-Toro, Tomás Arias-Vergara, Andreas Maier, Johann Jäger, Siming Bayer

    cs.LG · eess.SP

    The integration of renewable energy sources into the electrical grid introduces complex challenges in fault detection and coordination of grid recovery mechanisms. Traditional relay protection systems, which operate based on static rules and predefined thresholds, are inadequate for addressing these challenges, particularly in detecting and isolating faults such as short circuits. Consequently, the conventional methodologies applied to...

    arxiv.org/abs/2609.16744 · PDF

  40. 40

    Continuous-Time Machine Learning: A Unified Mathematical Perspective

    Waleed Razzaq, Yun-Sheng Zhao, Yun-Bo Zhao

    cs.LG · cs.AI

    Continuous-time (CT) machine learning has emerged as a principled framework for modeling temporal dynamics as a continuous process, particularly when observations are sampled at arbitrary time points or span long-range horizons. However, major branches of CT machine learning have matured in separate research communities, leaving their mathematical relationships and design trade-offs insufficiently characterized. In this survey, we develop a...

    arxiv.org/abs/2609.16710 · PDF

  41. 41

    Right Direction, Wrong Step: Geometric Analysis of Finite-Step Failure in Looped Transformers

    Zhihao Guo, Zonghan Wu, Haizhou Du, Huan Huo, Yilei Shao, Athanasios V. Vasilakos, Qingsong Wen

    cs.LG · cs.AI

    Looped Transformers offer a parameter-efficient route to test-time scaling by reusing shared layers for iterative latent reasoning. However, additional iterations can reduce support for a reference answer, leaving unclear whether an update's direction is locally unhelpful or its full displacement moves too far. We study this distinction by analysing reference utility, which measures this support, along the model's own update direction,...

    arxiv.org/abs/2609.16665 · PDF

  42. 42

    GrowMTP: Can RL Grow Its Own Draft Head?

    Minghua He, Lingzhe Zhang, Yuan Liu, Xiao Zhou, Aiwei Liu

    cs.LG · cs.CL

    Reinforcement learning (RL) post-training drives the frontier capabilities of large language models, with its wall-clock dominated by autoregressive rollout generation. Speculative decoding is an established remedy for this bottleneck, but existing draft heads must be pretrained or warmed up before RL, introducing substantial training cost outside the RL run to be accelerated. We observe that RL training itself provides both conditions...

    arxiv.org/abs/2609.16648 · PDF

  43. 43

    Stable by Construction: Variational Latent Markov Operators for Long-Horizon PDE Prediction

    Junyi Liao, Johann Guilleminot, Vahid Tarokh

    cs.LG · stat.ML

    Neural PDE solvers provide efficient surrogates for time-dependent physical systems, but autoregressive prediction over long horizons remains challenging because local errors can induce distribution shift and accumulate under recursive deployment. We develop a variational approach to this problem by introducing latent Markov dynamics in which physical states are represented by latent distributions and evolved through probabilistic...

    arxiv.org/abs/2609.16621 · PDF

  44. 44

    Divergence Timing and Cumulative Disagreement under KV-Cache Eviction

    Xinyue Luo, Fei Yu

    cs.LG

    KV-cache eviction perturbs the conditional token distributions governing autoregressive generation. We investigate how first-divergence timing and subsequent token mismatch determine cumulative disagreement. We derive an exact decomposition under a specified stepwise maximal coupling: the expected mismatch fraction equals a first-mismatch contribution plus post-divergence exposure multiplied by its mismatch rate. An explicit construction over...

    arxiv.org/abs/2609.16617 · PDF

  45. 45

    A Weighted Kernel Method for Approximation that Adapts to Learned Multivariable Structure

    John E. Darges, Laura Weidensager

    cs.LG · math.NA

    Approximating the input-output behavior of a multivariable black-box function from limited data is challenging when blind to the importance of its inputs and their interactions. We introduce total sensitivity kernels (TSKs), a method based on families of weighted ANOVA kernels that learn and adapt to this multivariable structure. TSKs parameterize the weights on each multivariable component of the target function by factors for each input. We...

    arxiv.org/abs/2609.16606 · PDF

  46. 46

    Recovering Physical Parameters from Fragmented Observations via Exact Distributed Spline Merging

    Naveen Mysore

    cs.LG · physics.comp-ph

    Scientific measurements are frequently distributed across locations, time periods, and institutions. Combining such fragments into a continuous, differentiable field enables recovering governing physical parameters from its derivatives. This paper makes two contributions toward that goal. First, the established additive structure of fixed-basis ridge-regression statistics is applied to tensor-product spline fields: each data holder computes a...

    arxiv.org/abs/2609.16579 · PDF

  47. 47

    AsyncCouple-Flow: Asynchronous Cross-Modal Coupling and Flow Matching for Spatio-Temporal Forecasting

    Zhixiang Wu, Yining Liu, Bo Zhao, Szu-Yu Chen, Huiran Duan, Chu Lin, Chuanguang Yang

    cs.LG

    Multi-modal spatio-temporal forecasting (MM-STF) supports weather nowcasting, traffic prediction, and earth-system modeling by combining heterogeneous sources such as physical fields, satellite imagery, and in-situ sensors. Three obstacles persist: (i) modalities have different spatio-temporal sampling rates, forcing lossy interpolation onto a unified grid; (ii) modalities are frequently missing at deployment due to sensor outages or revisit...

    arxiv.org/abs/2609.16573 · PDF

  48. 48

    On the Importance of Gating: Memorization vs. In-Context Learning in State Space Models

    William L. Tong, Aryo Lotfi, Emmanuel Abbe, Kostas Vaggelakos, Vishnu Banna, Etai Littwin, Josh Susskind, Cengiz...

    cs.LG · cs.AI

    State Space Models (SSMs) have emerged as a compelling alternative to Transformers, enabling sequence modeling with constant memory and linear compute. Although SSMs exhibit reasonable performance and favorable computational characteristics, they continue to lag behind Transformers on tasks that require in-context learning and precise retrieval, slowing their adoption for large-scale language modeling. In this work, we demonstrate that both...

    arxiv.org/abs/2609.16540 · PDF

  49. 49

    What Does Layer-Importance Reveal About Transformers and State-Space Models?

    Istabrak Abbes, Nizar Islah, Irina Rish, Sarath Chandar

    cs.LG · cs.AI

    Transformers and state-space models (SSMs) are the two dominant families of sequence models, and a central open question is how far the analytical knowledge built for transformers transfers to SSMs. We address this through the lens of layer importance which underpins compression, selective fine-tuning, and interpretability across both families. We decompose layer importance into two distinct notions. \emph{Necessity} captures how much the...

    arxiv.org/abs/2609.16537 · PDF

  50. 50

    FlowATC: Aircraft Trajectory Prediction via Flow Matching

    Mathurin Petit, Emir Torun, Louis Brusset, Jordan Kam, Alexandre M. Bayen

    cs.LG

    Building accurate decision-support tools for next-generation air traffic control requires robust trajectory prediction models. We present a flow-matching architecture trained exclusively on historical aircraft trajectories, with no route labels or chart supervision. Trained on 1.15 million Automatic Dependent Surveillance-Broadcast trajectory windows collected over the San Francisco Bay Area, the model generates aircraft trajectory...

    arxiv.org/abs/2609.16528 · PDF

  51. 51

    High-Performance Tensor Formulation of the Viterbi Algorithm for Hidden Semi-Markov Models

    Lorenzo Piarulli, Elia Belli, Daniele De Sensi

    cs.LG · cs.DC · cs.DS

    Hidden Semi-Markov Models (HSMMs) are fundamental probabilistic models widely adopted across diverse domains, from computational biology to finance and signal processing. The Viterbi algorithm decodes the most likely state sequence given an HSMM and can be applied iteratively for ab initio model learning. However, existing Viterbi implementations remain sequential, and GPU-accelerated solutions are entirely absent, making HSMM decoding...

    arxiv.org/abs/2609.16500 · PDF

  52. 52

    Decoder Design Matters for ECG Delineation

    Joseph Scharpf, William Han, Chaojing Duan, Michael A. Rosenberg, Emerson Liu, Ding Zhao

    cs.LG · cs.AI

    Electrocardiogram (ECG) delineation identifies the boundaries of P waves, QRS complexes, and T waves, providing structural annotations that can guide AI models in learning to interpret ECGs. However, training accurate delineation models requires manual annotations that are scarce and time-consuming to obtain. Recent work addresses this limitation through semi-supervised learning (SSL), but the design of the architecture, particularly the...

    arxiv.org/abs/2609.16489 · PDF

  53. 53

    Online Gradient Computation for Warping Gaussian Process Transformations

    Emilio Ruiz-Moreno, Konstantinos Slavakis, Baltasar Beferull-Lozano

    cs.LG · eess.SP

    Warped Gaussian processes (GPs) handle non-Gaussian observations by mapping them into a latent standard GP via a parametric transformation called warping. Existing streaming variants, however, either optimize the warping parameters periodically or sacrifice analytical tractability for a higher model capacity. To bridge this gap, we show that the gradient of the instantaneous negative log-likelihood of a warped GP admits an exact recursive...

    arxiv.org/abs/2609.16472 · PDF

  54. 54

    OPD-Aha: From Linguistic Momentum to Visual Reflection in Multimodal On-Policy Distillation

    Chenhao Qiu, Dawei Li, Yechao Zhang, Lei Gong, Zhen Tan

    cs.LG · cs.AI · cs.CV

    Privileged on-policy distillation improves multimodal reasoning by allowing a teacher to evaluate student trajectories using rich, training-only visual evidence. Both models score these trajectories while conditioning on the same student-generated prefix. When a student misinterprets an image early in a response, this accumulating erroneous rationale eventually pulls the teacher away from its visual evidence. The teacher and student converge...

    arxiv.org/abs/2609.16459 · PDF

This edition is part of The Daily Abstract — cs.LG archive. Subscribe to receive these in your inbox each morning, automatically translated to Spanish, with reply-to-PDF: arxivdaily.ignorelist.com.

Colophon Set in Georgia, with system sans for interface chrome and a monospaced stack for code and paper identifiers. Sole accent: amber #D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.