cs.LG · 2026-07-29 · No. 68

Machine Learning, 2026-07-29.

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

01 — The papers

52 entries
  1. 01

    Spend Experts Where You Are Unsure: Confidence-Adaptive Routing for Mixture-of-Experts LoRA

    Tom Saliencro, Rohan Desai, Priya Nair, Maya Lindqvist, Daniel Whitmore

    cs.LG

    Mixture-of-Experts (MoE) variants of Low-Rank Adaptation (LoRA) route every token to a fixed number of experts $k$. Tokens differ in how uncertain the model is about them, so a single k over-spends on easy tokens and under-serves hard ones. We observe that the router's output distribution is already a per-token uncertainty signal: peaked mass indicates confidence, while a flat distribution indicates ambiguity. We introduce CARE...

    arxiv.org/abs/2607.26052 · PDF

  2. 02

    Re-thinking Mammography Transfer Learning: The Dataset-Informed Transfer Learning (DITL) Framework for Breast Cancer Screening and Lesion Diagnosis

    Adarsh Bhandary Panambur, Siming Bayer, Andreas Maier

    cs.LG

    Enhancing classification performance in mammography remains a persistent challenge across both small curated datasets and large-scale clinical cohorts. Conventional transfer learning approaches often neglect dataset-specific characteristics, while recent neighborhood-informed methods have been restricted to narrow tasks with rigid formulations, limiting their scalability to population-level datasets. To address these challenges, we propose...

    arxiv.org/abs/2607.26043 · PDF

  3. 03

    Reinformed Dreamer: An Asymmetric World Model Efficiently Trained through Latent Guidance

    Gaspard Lambrechts, Adrien Bolland, Daniel Ebi, Damien Ernst

    cs.LG · stat.ML

    Much like humans benefit from guidance while learning, reinforcement learning algorithms may benefit from additional supervision beyond rewards. Leveraging additional information during training to learn better representations and behaviors has been the focus of asymmetric reinforcement learning. This learning paradigm has proven effective under partial observability when additional state information is available, but also under full...

    arxiv.org/abs/2607.26040 · PDF

  4. 04

    Sharpness-Aware Minimization and Muon: Robustness under the Spectral Norm

    Wenzhi Zhong, Edward Milsom, Michael Murray

    cs.LG · stat.ML

    Sharpness-Aware Minimization (SAM) aims to improve generalization by encouraging insensitivity to small, worst-case parameter perturbations. However, the notion of a "small" perturbation is inherently geometry-dependent: while existing SAM variants have explored a wide range of choices, a clear perspective on which geometries are most effective in practice remains elusive. Recent work on matrix-aware optimization, particularly the Muon...

    arxiv.org/abs/2607.26001 · PDF

  5. 05

    Empirical Evaluation of Out-Of-Distribution Performance of Tabular Foundation Models

    Malena Loza, David Chushig-Muzo, Eva Milara, Luis Bote-Curiel, Luis Estrada-Petrocelli, Felipe Grijalva

    cs.LG · cs.AI

    Tabular Foundation Models (TFMs) have emerged as novel approaches for tabular predictive tasks, demonstrating competitive predictive performance to ensemble tree-based models. Most TFMs are trained and evaluated on independent and identically distributed data, but this assumption changes in real-world scenarios due to distribution shifts, which compromise the robustness of models. Limited research has been conducted of TFMs under distribution...

    arxiv.org/abs/2607.26000 · PDF

  6. 06

    Generator-Aligned Representation Interfaces for Diagnostic Soft Equivariance

    Weitao Li, Gong Cheng

    cs.LG

    Exact-equivariant architectures typically encode prescribed group actions in specialized operators, which can complicate their reuse with generic backbones and across data modalities. We introduce the Generator-Aligned Representation Interface (GARI), a representation-level design principle that exposes selected transformation generators to a generic sequence backbone through aligned canonical and generator-induced views. We formalize the...

    arxiv.org/abs/2607.25988 · PDF

  7. 07

    Reinforcement Learning for Code Optimization

    Pierre Chambon, Kunhao Zheng, Juliette Decugis, Benoit Sagot, Gabriel Synnaeve

    cs.LG · cs.AI

    RL for code correctness is now established: have the model generate a program, run it against hidden test cases, and reward solutions that pass. Extending this to code optimization seems straightforward: just add execution time to the reward. But in practice, once timing drives the reward, small problems in measurement noise, reward sparsity, or GRPO instability overwhelm the signal and make RL fail: generated solutions are barely faster, and...

    arxiv.org/abs/2607.25970 · PDF

  8. 08

    Minimizing Targeted Activations: Input-Only Suppression of Evaluation-Awareness Latents in Large Language Models

    Deepanshu Mody, Samarth Agarwal, Utkarsh Mittal, Dipesh Mahato

    cs.LG · cs.AI · cs.CL

    Activation steering controls model behavior by editing internal activations at inference time. We study its input-side dual: optimizing a fluent prompt so that a chosen internal latent is driven toward zero, with no inference-time model access. Our target is an "evaluation-awareness" latent-linearly readable and steerable in recent work-whose control would threaten the validity of safety evaluations if models behave differently when they...

    arxiv.org/abs/2607.25907 · PDF

  9. 09

    A Machine-Learning-Based Gas Lift Optimization Workflow for Unconventional Fields

    Sha, Miao, Alexandra Vendetti, Logan Smart, Gunta Chomchalerm, Yang Chen, Christopher Frazier, Dustin Haralson,...

    cs.LG · cs.AI · cs.SE

    In this paper, we present an automated data-driven workflow using Machine Learning (ML) for gas lift optimization in unconventional fields. This workflow integrates a ML model that accurately forecasts the Gas Lift Performance Curve, and a Bayesian Optimization Framework to solve for the optimal gas injection rates under the constraints of facility capacity. The ML model leverages the historical production time series data without requiring...

    arxiv.org/abs/2607.25885 · PDF

  10. 10

    A2TTA: Anchored-and-Agile Test-Time Adaptation for Evolving Traffic Sensor Networks

    Du Yin, Xiachong Lin, Yue Tan, Jinliang Deng, Estrid He, Hao Xue, Flora D. Salim

    cs.LG · cs.AI

    Traffic forecasting is important for efficient traffic management and route planning in smart cities. Existing traffic forecasting studies typically assume fixed sensor graphs, overlooking the continuous evolution of real-world traffic networks, e.g., ongoing road network construction and evolving human mobility patterns. These dynamic changes can substantially degrade conventional forecasting models, motivating test-time adaptation (TTA) to...

    arxiv.org/abs/2607.25875 · PDF

  11. 11

    DRIFT: Direct-Recursive Intervention-Conditioned Forecasting of ICU Physiological Trajectories

    Weixin Liu, Juming Xiong, Congning Ni, Yanfan Zhu, Xingtao Lin, Bradley A. Malin, Zhijun Yin

    cs.LG · eess.SP

    Many time-series forecasts depend not only on prior observations but also on actions specified during the forecast period. In intensive care units (ICUs), future vital signs and laboratory values are influenced by treatments such as vasopressors. However, models that predict the full future sequence all at once make little use of these treatments, whereas autoregressive models can accumulate errors. We introduce DRIFT, a hybrid framework in...

    arxiv.org/abs/2607.25864 · PDF

  12. 12

    Prototype Adaptation for Zero-Shot sEMG Movement Classification

    Rui Liu, Benjamin Paassen

    cs.LG

    Surface electromyography (sEMG) enables the control of prostheses, allowing upper-limb amputees to re-gain some hand function. Most current research focuses on recognizing basic movements for prosthesis control. However, in most daily activities, such as opening a door, combined movements are essential. However, collecting training data for all possible combined movements is time-consuming and requires re-training of the model for any new...

    arxiv.org/abs/2607.25826 · PDF

  13. 13

    SpectONet: A Physics-Guided Spectral Deep Operator Network for Euler-Bernoulli Beam Dynamics

    Shivani Saini, Ramesh Kumar Vats, Arup Kumar Sahoo

    cs.LG · cs.AI · math.DS

    This paper proposes a novel physics-guided spectral deep operator network, termed SpectONet, for solving Euler-Bernoulli beam (EBB) vibration problems. The proposed framework integrates the operator-learning capability of DeepONet with physics-informed constraints and Chebyshev-Gauss-Lobatto (CGL) sensor placement. Unlike conventional DeepONet frameworks, which commonly employ uniformly distributed sensors, SpectONet uses nonuniform spectral...

    arxiv.org/abs/2607.25790 · PDF

  14. 14

    An Embarrassingly Simple Rule-based Visiting Circulation Approach to Trip Destination Prediction

    Eng-Shen Tu, Yong-Han Chen, En-Chao Liu, Hao-Yun Keng, Cheng-Te Li

    cs.LG

    In this paper, we propose the Rule-based Visiting Circulation (RVC) model in tackling the challenge in the IEEE Big Data Cup 2022: Trip Destination Prediction. Given trips containing travel information, personal attributes, origin zones, and their features in the training metropolitan areas, the task is to predict the destination of every trip in a targeted metropolitan area whose destinations are not given at all at the training stage. We...

    arxiv.org/abs/2607.25751 · PDF

  15. 15

    Detecting CSAM Text-to-Image LoRAs From Weights

    David Demitri Africa, Cate Heine, Nadine Staes-Polet, Kimberly Mai

    cs.LG · cs.CY

    Low-rank adaptation (LoRA) fine-tuning has made it cheap and easy to customize open-weight image generation models for specific tasks, including the production of child sexual abuse material (CSAM). Existing moderation relies on metadata or generated outputs, but metadata can be deceptive and generating outputs may itself be unacceptable or illegal. We show that a safer signal lives in the weights. The top-left singular vectors of a LoRA's...

    arxiv.org/abs/2607.25750 · PDF

  16. 16

    Optimization with Dynamic Constraint Learning (DCL)

    Ezgi Oztekin, Figen Oztoprak, S. Ilker Birbil

    cs.LG · math.OC

    We propose Dynamic Constraint Learning (DCL), a data-driven framework for constrained optimization when constraint functions are unknown and cannot be queried during optimization. At each iteration, the method learns a local surrogate from nearby data and solves a subproblem within a data-supported trust region. Compared with offline global constraint learning, the approach uses local surrogates that adapt to the data distribution during...

    arxiv.org/abs/2607.25719 · PDF

  17. 17

    Tools Are Not Islands: Set-Level Tool Retrieval for LLM Agents via Query-Conditioned Hyperedge Prediction

    Xinyi Hong, Pinjun Dong, Xinyang Yu, Binyan Jiang

    cs.LG · cs.AI · cs.IR

    Large language model (LLM) agents increasingly rely on invoking external tools to complete real-world tasks. Tool retrieval, which selects a small task-relevant subset from a library of thousands of tools before the agent acts, has therefore become a critical component of LLM agent pipelines. However, existing retrievers either score each tool in isolation or assemble the tool set sequentially, so the joint utility of a candidate set is never...

    arxiv.org/abs/2607.25718 · PDF

  18. 18

    From Deterministic to Generative Deep Learning for Urban Air Quality Reconstruction from Sparse Observations

    Abhishek A. Sabnis, Mihai Mitrea, Lya Lugon, Karine Sartelet, Marc Bocquet, Xiaoyuan Cheng, Shupeng Zhu, Sibo Cheng

    cs.LG · cs.AI

    Full-field reconstruction of air pollution is essential for evaluating pollution exposure and supporting public health decision-making. However, the complex interactions among pollutants, hard-to-predict weather patterns, and limited monitoring station coverage make this a complex task. We apply deep learning techniques to provide fast and accurate reconstructions from sparse observations of four key pollutants: NO2, O3, PM2.5 and PM10....

    arxiv.org/abs/2607.25687 · PDF

  19. 19

    Rashomon Alignment

    Moisés Santos, Peter van der Putten, Bernhard Pfahringer, Carlos Soares

    cs.LG · cs.AI

    We propose Rashomon Alignment (RA), a new measure to assess functional similarity between two models. Existing functional similarity measures are distributional, quantifying differences between outputs of models applied to real-world data. However, these measures can be regarded as ecologically valid only for regions in the input space represented by the available data. We introduce a geometrical perspective on functional model similarity,...

    arxiv.org/abs/2607.25680 · PDF

  20. 20

    DynaBridge: Dynamic Summary-Guided Cross-Task Multimodal Fusion for DASS-Structured Mental Health Assessment

    Shiyu Teng, Haichen Yu, Jiaqing Liu, Hao Sun, Yu Song, Shurong Chai, Ruibo Hou, Lanfen Lin, Yen-Wei Chen

    cs.LG · cs.AI · cs.MM

    Multimodal behavioral analysis offers a scalable approach to assessing depression, anxiety, and stress, yet generic fusion models often ignore the psychometric structure of questionnaire labels. In DASS-21, risk labels are derived from ordered symptom items through fixed item-to-subscale mappings. We propose \textbf{DynaBridge}, a dynamic summary-guided cross-task multimodal framework for DASS-structured mental health assessment. DynaBridge...

    arxiv.org/abs/2607.25679 · PDF

  21. 21

    A Physics-Informed Neural Operator for Thermal Ranking of Low-Cost Wall Materials in Hot-Dry Climates

    Muhammad Akbar Khan, Fahim Raees, Ubaida Fatima

    cs.LG · math.NA · physics.comp-ph

    Identifying cost-effective indigenous building materials that minimise heat penetration through walls is critical for indoor thermal comfort in low-income rural housing in hot-dry climates, where summer temperatures routinely exceed 45 C. We present a two-stage computational framework for thermal ranking of five low-cost indigenous wall materials: mud brick, clay-straw adobe, lime-stabilised bamboo panel, fired clay brick, and lime-mud...

    arxiv.org/abs/2607.25668 · PDF

  22. 22

    Contextual Deconvolution for Variance-Stable Demand Sensing: Kernel-Modulated Operators in Promotional Retail

    Mohammad Forouhesh

    cs.LG · math.OC · stat.ML

    Machine learning demand forecasts optimize statistical accuracy yet leave excess operational volatility that inflates safety stock and amplifies the Bullwhip effect. We introduce \textbf{Contextual Deconvolution} (CD), a two-stage estimator that reframes demand sensing as a convex decomposition: a kernel-modulated banded operator separates transient promotion-driven shocks from a smooth structural baseline, and hierarchical partial pooling...

    arxiv.org/abs/2607.25664 · PDF

  23. 23

    Using Data-Derived Priors to Guide CNN Architecture Design for NIR Chemometrics

    Dário Passos

    cs.LG · physics.app-ph · physics.comp-ph

    Convolutional neural networks (CNN) for near-infrared (NIR) chemometrics are often designed using generic architectural rules, although spectral datasets differ in sampling, smoothness, redundancy, and sample size. We tested whether these properties can provide empirical priors for CNN design. Across 25 NIR regression tasks, we computed descriptors of dataset size, spectral length and spacing, entropy, intrinsic rank, autocorrelation, and...

    arxiv.org/abs/2607.25636 · PDF

  24. 24

    MemSFT: Mitigating Alignment Tax with an External Parametric Memory

    Jiarui Wang, Xiang Shi, Jiaqi Cao, Rubin Wei, Xiquan Wang, Hao Sun, Jingzhi Wang, Zhiqi Yang, Qipeng Guo, Bowen...

    cs.LG · cs.CL

    Adapting Large Language Models (LLMs) to specialized domains often incurs an alignment tax, as fine-tuning on domain-specific tasks can cause catastrophic forgetting and substantially degrade performance on general tasks. We propose MemSFT, which mitigates the alignment tax by decoupling domain specialization from backbone parameter updates through a plug-and-play parametric memory. The memory is trained to imitate the behavior of a...

    arxiv.org/abs/2607.25614 · PDF

  25. 25

    Contrastive Representation Learning of Longitudinal Disease Trajectories on Temporal Graphs

    Bastian Pfeifer

    cs.LG · cs.AI · q-bio.QM

    Understanding disease trajectories from longitudinal clinical data remains challenging due to complex temporal dynamics and heterogeneous patient cohorts. Here, we present a contrastive representation learning framework that models multivariate disease trajectories as temporal graphs and learns representations using contrastive graph neural networks. Nodes represent patient observations over time, while edges capture temporal continuity and...

    arxiv.org/abs/2607.25609 · PDF

  26. 26

    Physics-Informed Broad Learning System: An Efficient Backpropagation-Free Framework for Solving Partial Differential Equations

    Pinki Khatun, M. Sajid, Abhinav Jha, M. Tanveer

    cs.LG · cs.AI

    Physics-informed neural networks (PINNs) have emerged as a powerful paradigm for solving partial differential equations (PDEs) by embedding governing physical laws into deep neural networks. However, their reliance on computationally expensive gradient-based optimization and deep architectures often results in slow training, high computational cost, and limited scalability. In this work, we propose a novel physics-informed broad learning...

    arxiv.org/abs/2607.25608 · PDF

  27. 27

    OrthKD: Extracting Generalized Clinical Knowledge from Heterogeneous Teachers for Lightweight Deployment

    Yi Xu, Cheng Chen, Mufan Cao

    cs.LG · cs.CV

    Deploying diabetic retinopathy (DR) screening models in primary care requires edge-efficient systems that remain accurate, safe, and reliable under domain shift. Multi-teacher knowledge distillation (KD) is a natural compression strategy, but existing approaches largely assume that all teachers provide equally trustworthy supervision. In our setting, this assumption fails: a strong CNN teacher (EfficientNet-B3, 0.876 QWK) and a weaker...

    arxiv.org/abs/2607.25545 · PDF

  28. 28

    Mind the Missing Split: Resolving Feature Heterogeneity in Swarm Learning with Random Forests

    Mohammad Tajabadi, Dominik Heider

    cs.LG

    Swarm Learning is a decentralized collaborative learning mechanism that allows multiple organizations to train a shared model without central coordination or direct data sharing. In typical horizontal Swarm Learning, datasets across sites are usually assumed to share the same feature set. However, in real-world applications, sites often have partially overlapping features because measurements, protocols, and available covariates differ across...

    arxiv.org/abs/2607.25538 · PDF

  29. 29

    Multi-Scale Structural Features for Continual, Comprehensible Visual Recognition in a Developmental Learning Framework

    Zeki Doruk Erden

    cs.LG · cs.AI · cs.CV

    Contemporary machine learning struggles to learn continually, reuse prior knowledge, and expose a comprehensible internal structure. A recently proposed developmental, gradient-free learning framework addresses these limitations by learning a discrete, topological model of its inputs through local variation and selection, yielding an inherent continual-learning guarantee: new observations refine existing structure without overwriting past...

    arxiv.org/abs/2607.25531 · PDF

  30. 30

    AMPBench-MT: A Homology-Controlled Benchmark for Antimicrobial Peptide Potency, Spectrum, and Safety Prediction

    Ziheng Zhou, Huiyu Luo, Xiaohu Zhu, Nan Wang, Xuebiao Qin, Chaoyan Zhang, Jun Yan

    cs.LG · q-bio.QM

    Computational AMP discovery is often evaluated through AMP/non-AMP recognition, yet follow-up decisions depend on assay-derived evidence such as target-species potency, hemolysis, toxicity, and selectivity. Existing AMP and peptide benchmarks cover binary recognition, multilabel annotation, assay regression, or broader peptide-model comparison, but they do not jointly place AMP recognition, species-conditioned potency, spectrum, safety-facing...

    arxiv.org/abs/2607.25518 · PDF

  31. 31

    Anti-Backdoor Coreset Selection via Cumulative Entropy

    Qi Zhao, Christian Wressnegger

    cs.LG · cs.CR

    Recent training-time defenses against neural backdoors isolate a benign subset from poisoned training data, to learn a backdoor-free model from it. In this paper, we formulate this defense strategy as a coreset selection problem, giving rise to so-called "Anti-Backdoor Coreset Selection." Since poisonous samples have (a) lower prediction uncertainty and are (b) less frequent than benign samples, coreset selection naturally focuses more on...

    arxiv.org/abs/2607.25502 · PDF

  32. 32

    Quantum Speedups for Stochastic Optimization with Heavy-Tailed Noise

    Bin Luo, Chengchang Liu, Jonathan Allcock, Shengyu Zhang, John C. S. Lui

    cs.LG

    We study stochastic optimization with heavy-tailed gradient noise. We first propose a novel quantum mean estimator for multivariate heavy-tailed random variables that achieves lower query complexity than optimal classical estimators in the low-dimensional regime. We further develop an unbiased quantum mean estimator by applying a generalized multi-level Monte Carlo technique. We prove quantum lower bounds showing that, when the dimension $d$...

    arxiv.org/abs/2607.25492 · PDF

  33. 33

    Data-Dependent Regret and Polyak Corrections for Constrained Online Convex Optimization

    Wentao Zhang

    cs.LG

    Constrained online convex optimization requires minimizing regret against adversarial convex costs while satisfying a convex constraint at every round, as needed in safety-critical applications. A computationally efficient method combines online gradient descent with a Polyak feasibility step, using one constraint evaluation and one subgradient per round. Although this method achieves O(sqrt(T)) regret with per-round feasibility, we derive a...

    arxiv.org/abs/2607.25480 · PDF

  34. 34

    Emergent Latent-State Computation under Stochastic Volatility

    Xiaoyu Huang, Lulu Wang

    cs.LG · cs.AI · q-fin.ST

    Mechanistic interpretability has largely focused on language models and deterministic toy tasks. Much less is known about how sequence models internally represent latent stochastic dynamics under noisy, partially observed observations. We study this question in a controlled multivariate stochastic volatility setting, where models observe only returns while the ground-truth latent volatility state is known to the researcher. This setting...

    arxiv.org/abs/2607.25459 · PDF

  35. 35

    Bits and Memories: Measuring Verbatim Extraction Across LLM Quantization

    Akshay Sasi

    cs.LG · cs.CR

    Language models are almost always quantized before they are deployed, and a growing line of work asks whether quantization also lowers their privacy risk. That work measures privacy almost entirely with membership inference. We think this is the wrong thing to measure for the risk that most people actually worry about, namely a model reproducing its training data word for word, and we measure that directly. Using the Pythia models and the...

    arxiv.org/abs/2607.25451 · PDF

  36. 36

    PIcsC: Partitioning-Induced Covariate Shift Correction

    Behraj Khan, Behroz Mirza, Syed Ahmad Chan Bukhari, Tahir Qasim Syed

    cs.LG

    Covariate shift across training-data partitions biases model selection and parameter estimation in cross-validation, lifelong learning, and federated learning. We propose \textit{Partition-Induced Covariate-shift Correction} (\texttt{PIcsC}), a Fisher information-based regularization framework that mitigates distribution mismatch between data partitions and a reference distribution. \texttt{PIcsC} approximates partition divergence using the...

    arxiv.org/abs/2607.25441 · PDF

  37. 37

    SPARC Segmentation to Prediction via Affine Regression and Counterfactuals

    Shivani, Subhayan Roy

    cs.LG

    Transaction propensity prediction in B2B e commerce presents unique challenges distinct from B2C contexts, primarily due to the heterogeneous procurement behaviors of organizational entities, which violate SMOTE's implicit assumption of within class feature homogeneity. Specifically, B2B buyers exhibit multi modal procurement cycles that render linear interpolation between minority class samples structurally invalid, producing synthetic data...

    arxiv.org/abs/2607.25413 · PDF

  38. 38

    TWICE: Two-Clock, Two-Window Learning for Long-Horizon Conversion Prediction in Online Advertising

    Kaiyuan Li, Kun Wang, Zhongbo Wang, Teng Sha, Ming Yan, Yanhua Cheng, Xialong Liu

    cs.LG · cs.IR

    Long-horizon conversion prediction under delayed feedback creates a two-clock, two-window learning problem in online advertising. A short base observation window releases recent clicks on the click clock before their outcomes mature, whereas conversions continue to arrive on the conversion clock throughout a longer target conversion window. The click clock provides timely but partially observed status supervision. The conversion clock reveals...

    arxiv.org/abs/2607.25404 · PDF

  39. 39

    Learned, Relied Upon, or Necessary? Separating Checkpoint Dependence from Task-Level Value in Sheaf GNNs

    Yi Liu

    cs.LG

    Learned restriction maps in sheaf graph neural networks are often treated as proof that the model has discovered useful edge geometry. That conclusion does not follow from parameter movement or from a post-hoc ablation: both can show how one checkpoint is organized while leaving open whether learned transport still helps after the rest of the model adapts. We separate these claims with two estimands. Checkpoint reliance intervenes on the maps...

    arxiv.org/abs/2607.25387 · PDF

  40. 40

    Rethinking Likelihood distributions: Student's t Likelihood Boosts Bayesian Neural Network Performance

    Pei-Hsuan Hsia, Lars H. Heyen, Arvid Weyrauch, Markus Goetz, Achim Streit, Sebastian Krumscheid, Charlotte Debus

    cs.LG · cs.AI

    In Bayesian neural networks (BNNs), variational inference is a widely adopted framework for modeling uncertainty in a distributional way, with the evidence lower bound (ELBO) serving as the standard objective function. Several distributions contribute to the ELBO loss, such as the prior, approximated posterior, and likelihood distribution. Typically, these distributions are all approximated by a Gaussian distribution, since it is easy to...

    arxiv.org/abs/2607.25376 · PDF

  41. 41

    Raven: High-Recall Sequence Modeling with Sparse Memory Routing

    Arshia Afzal, Aviv Bick, Eric P. Xing, Volkan Cevher, Albert Gu

    cs.LG · cs.AI

    Long-context recall in linear-time sequence models highlights a tradeoff in how they write to memory. State-based linear models, such as state-space models (SSMs) and linear Transformers, write densely, updating the entire state for each newly arrived token, which leads to interference and makes specific past tokens hard to recover. Sliding-window attention (SWA) exhibits the opposite behavior: it writes sparsely by storing explicit token...

    arxiv.org/abs/2607.25357 · PDF

  42. 42

    Explainable AI for Chronic Kidney Disease Prediction Using Simulated Federated Learning

    Md Zahid Hasan Ontor, Md Al Amin, Anik Dev Nath, Bikash Kumar Paul

    cs.LG · cs.AI

    Chronic Kidney Disease (CKD), characterized by the gradual loss of kidney function, remains a significant public health challenge. Early detection is crucial for preventing severe complications and enhancing patient outcomes. In this study, Federated Learning (FL) with a VotingClassifier was used to predict CKD using a clinical dataset, where Random Forest, AdaBoost, and XGBoost were utilized to compare and identify the best-fitting model for...

    arxiv.org/abs/2607.25348 · PDF

  43. 43

    Guiding Posterior Exploration with Optimizer-Derived Geometry

    Moritz Schlager, Emanuel Sommer, Thomas Möllenhoff, David Rügamer

    cs.LG

    Sampling-based methods offer a principled approach to uncertainty quantification in Bayesian neural networks. Their practical use, however, is often challenged by the computational cost of exploring high-dimensional and multimodal posterior distributions. To overcome these difficulties, Bayesian Deep Ensembles, i.e., warmstarting the sampling from several optimized solutions, have proven to be an effective strategy. In this paper, we...

    arxiv.org/abs/2607.25312 · PDF

  44. 44

    Retraction-Free Optimization over the Stiefel Manifold for the LoRA Fine-Tuning

    Yuan Zhang, Jiang Hu, Zhijian Lai, Lin Lin, Zaiwen Wen

    cs.LG · cs.AI

    Optimization over the Stiefel manifold plays a significant role in various machine learning tasks. Existing methods either use the retraction operators, requiring costly orthonormalization for large-scale matrices, or employ landing methods that rely on careful step size selection and penalty parameter tuning. To address these challenges, we propose a retraction-free and penalty parameter-free algorithm that directly lands on the manifold. By...

    arxiv.org/abs/2607.25299 · PDF

  45. 45

    Zhinv: Real-time hub-height wind field reconstruction using only local sparse observations

    Zongwei Zhang, Chin Chun Ooi, Lianlei Lin, Sheng Gao, Tiantian He, Yew Soon Ong, Junkai Wang, Hangyi Yu, Jiaqi...

    cs.LG

    The high proportion of wind power connected to the grid places higher demands on fine-grained knowledge of regional wind fields. Since the wind information directly obtainable in actual operations is mostly sparse, discrete, and irregularly distributed local observations, it is difficult to directly meet the needs of tasks such as wind power regulation, wind resource assessment, and low-altitude environmental perception of continuous regional...

    arxiv.org/abs/2607.25298 · PDF

  46. 46

    Breaking the Periodicity Assumption: Robust Tensorial Multi-View Clustering via Graph-Spectral Low-Rank Learning

    Jintian Ji, Xingsu Li, Songhe Feng

    cs.LG

    Tensorial multi-view clustering (TMC) has achieved strong performance due to its ability to capture high-order correlations across multiple views. Most existing t-SVD-based TMC frameworks apply the Fast Fourier Transform (FFT) along the sample mode to impose frequency-domain low-rank constraints. However, we reveal that this widely adopted design critically relies on an implicit ``periodicity assumption'' induced by the sample arrangement....

    arxiv.org/abs/2607.25295 · PDF

  47. 47

    AMRD: Adaptive Multi-Teacher Relational Distillation for Lightweight Speech Emotion Recognition

    Yuqi Li, Yi-Cheng Lin, Xianglong Wang, Kuo Yang, Xiaoqin Feng, Yixuan Wang, Huiran Duan, Yingli Tian

    cs.LG

    On-device speech emotion recognition (SER) is critical for real-time applications, yet large self-supervised models that excel at SER are too costly for edge devices. Multi-teacher knowledge distillation can compress them into a lightweight student, but two challenges remain: teacher reliability varies across batches, and logit-level distillation ignores inter-sample relational structure. We propose Adaptive Multi-teacher Relational...

    arxiv.org/abs/2607.25289 · PDF

  48. 48

    When Does Deep Representation Learning Help Single-Cell Clustering? A Sensitivity-Aware Diagnostic Benchmark for Biomedical AI Pipelines

    Nguyen Thanh Phong, Truong Viet Vu, Nguyen Ha Thu, Tran An Ky, Tran Hoang Thong, Le Pham Thuy Hien, Nguyen Thai Anh

    cs.LG · q-bio.GN

    Single-cell ribonucleic acid sequencing (scRNA-seq) is a foundational technology for precision-medicine workflows that contribute to United Nations Sustainable Development Goal 3 on Good Health and Well-being, and unsupervised clustering is the analytical step that turns raw expression matrices into interpretable cell populations. Practitioners therefore face a recurring engineering decision: is an additional deep representation stage worth...

    arxiv.org/abs/2607.25288 · PDF

  49. 49

    HeAD-CP: Heterophily-Aware Diffused Conformal Prediction Sets for Graph Neural Networks

    Phan Binh Nguyen Lam, Nguyen Thai Anh

    cs.LG

    Conformal prediction (CP) provides distribution-free uncertainty quantification, and its extension to graphs is an active research direction. Diffused Adaptive Prediction Sets (DAPS) is a widely used graph-aware diffusion baseline, propagating Adaptive Prediction Sets (APS) non-conformity scores along edges with a uniform coefficient $λ$. We identify a fundamental shortcoming of this design: the uniform low-pass diffusion presupposes graph...

    arxiv.org/abs/2607.25273 · PDF

  50. 50

    Bridging Compute- and Data-Optimal Pretraining

    Tian Qin, Kimia Hamidieh, David Alvarez-Melis

    cs.LG · cs.AI · cs.PF

    Classical compute-optimal scaling laws assume an unbounded supply of fresh pretraining data, yet pretraining is increasingly entering a regime in which compute grows faster than the availability of high-quality data. We propose Compute-Data (CD) scaling laws, a unified framework that bridges compute-optimal scaling, where data scales freely with compute, and data-optimal scaling, where the corpus is fixed while compute can grow without bound....

    arxiv.org/abs/2607.25271 · PDF

  51. 51

    Beyond Single-Episode Optimization: Sliding-Window Aware Generative Auto-Bidding for Long-Term Advertising Effectiveness

    Binglin Wu, Chuan Yue, Yingyi Zhang, Xianneng Li, Ruyue Deng, Weiru Zhang, Xiaoyi Zeng

    cs.LG

    Auto-bidding systems optimize bids to maximize value under efficiency constraints such as Cost-Per-Action (CPA). Existing methods treat each day as an independent episode. However, many advertisers produce value so sparsely that per-day efficiency ratios become statistically unreliable, undermining advertiser retention. Platforms therefore evaluate window-level efficiency over sliding windows of $W{=}7$ days, ensuring fair evaluation and...

    arxiv.org/abs/2607.25233 · PDF

  52. 52

    Neurai-VN Benchmark: Standardized Machine Learning Models for Multimodal Digital Phenotyping in Mental Health Classification

    Quoc-Cuong Pham, Hoang-Thuy-Duong Vu, Thi-Thanh-Huong Ha, Huy-Hieu Pham

    cs.LG

    Digital phenotyping (DP) using smartphones and wearable devices has shown considerable potential for mental health monitoring. However, progress remains difficult to evaluate due to heterogeneous datasets, inconsistent preprocessing pipelines. In this study, we present a reproducible benchmark built upon the Neurai-VN dataset, a high-resolution, multimodal dataset comprising passive sensing and active assessment from wearable and smartphone...

    arxiv.org/abs/2607.25232 · 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.