cs.LG · 2026-09-21 · No. 120

Machine Learning, 2026-09-21.

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

01 — The papers

60 entries
  1. 01

    BrainWideBench: Benchmarking large-scale pretraining and across-animal transfer in multi-region neural recordings

    Alexandre Andre, Shivashriganesh P. Mahato, Vinam Arora, Keshav Balaji, Divyansha Lachi, Nanda H. Krishna, Jingyun...

    cs.LG · q-bio.NC

    Advances in large-scale neural recording have made it possible to collect data across many animals and distributed brain regions, raising the question of whether this scale can be exploited to learn general-purpose neural representations transferable across diverse downstream tasks. Yet, progress toward this goal has been limited by fragmented evaluation protocols and a narrow focus on individual task domains. Here, we present BrainWideBench,...

    arxiv.org/abs/2609.22064 · PDF

  2. 02

    Benchmarking World Models for Continual Learning on Compositional Tasks

    Haoyu Zhou, Joe Watson, Anson Lei, Ingmar Posner

    cs.LG · cs.RO

    A desirable property of a world model is the ability to learn continually across tasks, adapting to new environments without forgetting what the agent has already learnt. In particular, the ability to retain and reuse knowledge obtained from prior experiences underpins an agent's ability to efficiently adapt to novel environments, as the dynamics of the physical world can often be described in recurring mechanisms. However, the world model's...

    arxiv.org/abs/2609.22055 · PDF

  3. 03

    Particle Competition and Cooperation for Robust Graph Convolutional Network Learning Under Label Noise

    Fabricio Breve

    cs.LG

    Graph Convolutional Networks (GCNs) are highly sensitive to label noise, since corrupted supervision can propagate through the graph and degrade learned node representations. This work proposes PCC+GCN, a hybrid framework that uses Particle Competition and Cooperation (PCC) as a graph-based label-refinement stage before GCN training. PCC identifies suspicious labeled nodes through particle domination dynamics and determines whether their...

    arxiv.org/abs/2609.22053 · PDF

  4. 04

    Available Guardrails: Certifying Selective Prediction across ML Systems

    Parivesh Priye, Yufeng Wang, Haibin Ling, Michael Chaykowsky

    cs.LG

    A selective predictor acts as a safety gate: it returns an output only when the prediction appears sufficiently trustworthy. Deployments increasingly require this reliability to be certified at a target precision for every reporting unit of interest, such as a tool, policy label, or patient subgroup. The main difficulty is often not whether a granted certificate is valid, but whether finite calibration data can produce one at all. As the gate...

    arxiv.org/abs/2609.22048 · PDF

  5. 05

    $λ$-Controlled GRPO: Turning Flow-Matching Ratio Instability into a Budgeted Resource

    Yufeng Wang, Parivesh Priye, Meeshawn Marathe, Ramit Pahwa

    cs.LG

    Reinforcement learning is increasingly used to align image generators with reward signals, and Flow-GRPO recently extended this paradigm to flow-matching models by treating the denoising sampler as a stochastic policy that can be optimized from reward feedback. Training in this setting is unstable in a way specific to multi-step denoising: the policy update changes systematically across denoising steps, with importance ratios drifting below...

    arxiv.org/abs/2609.22041 · PDF

  6. 06

    COMPLEX: A Closed-Form Certified Embedding of Multiparameter Persistence Modules

    Sushovan Majhi, Atish Mitra, Žiga Virk, Pramita Bagchi

    cs.LG · math.AT

    Every multiparameter persistence vectorization we know of carries a one-sided Lipschitz upper bound and nothing below it: without a lower gauge there is no sense in which the features are faithful, and no per-prediction guarantee can be built on them. This paper supplies the missing side. 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...

    arxiv.org/abs/2609.22012 · PDF

  7. 07

    Abstention and Noise Filtering: Two Missing Primitives of Softmax Attention

    Richard Zhe Wang

    cs.LG · cs.CL

    Gating the value pathway of attention reportedly improves language model pretraining, and prior studies disagree on why. We argue and provide experimental evidence that such gates supply two different things that softmax attention lacks: abstention and noise filtering. The first is abstention, which allows an attention head to output nothing, bypassing the requirement that attention weights must sum to one. The second is noise filtering,...

    arxiv.org/abs/2609.22005 · PDF

  8. 08

    Assessment of Machine Learning-Based Critical Heat Flux Models in the CTF Subchannel Code for Square Rod Bundle Prediction

    Aidan Furlong, Vinicius de Melo Monteiro, Robert Salko, Juliana Pacheco Duarte, Xu Wu

    cs.LG

    The prediction of critical heat flux (CHF), a key safety-related quantity in nuclear thermal hydraulics, remains an important challenge due to its direct relationship with fuel performance and reactor safety. Recent studies have demonstrated that relative to traditional empirical correlations and lookup tables (LUTs), machine learning (ML) methods can substantially improve CHF prediction accuracy. Most ML-based CHF models, however, have been...

    arxiv.org/abs/2609.21995 · PDF

  9. 09

    Time series generation with spectrally aligned latent flow matching

    Camilo Carvajal Reyes, Felipe Tobar

    cs.LG

    Latent flow models have proven to be a reliable and cost-effective method for time series generation. However, the latent compression induces unwanted artefacts, such as a spectral mismatch with respect to the underlying dataset, thus hindering their use as training surrogates. In this article, we propose a spectrally-aligned latent-flow time series generator, where the latent space for flow matching is trained to preserve dynamical...

    arxiv.org/abs/2609.21989 · PDF

  10. 10

    RACER: Role-Aligned Competence Estimation for Human-AI Routing

    Joshua Strong, Emma Sun, Alexander Capstick, Pramit Saha, Cheng Ouyang, J. Alison Noble

    cs.LG

    Learning to defer asks a predictive system when to act autonomously and when to defer to a human expert. Population-adaptive deferral extends this problem to unseen experts using a small context set of expert behavior. Neural context encoders such as L2D-Pop can be query-dependent, but may learn routing shortcuts tied to absolute class coordinates. Identity-Free Deferral (IFD) removes such shortcuts through role-indexed classwise competence...

    arxiv.org/abs/2609.21953 · PDF

  11. 11

    Learning to Move Cities: Deep Meta-Models and Reinforcement Policies for Calibration and Control in Urban Networks

    Adewumi Augustine Adepitan, Christopher J. Haruna, Oluwasegun Adegoke, Ayooluwatomiwa Ajiboye, Oluwatobi Oluwasakin

    cs.LG

    Urban transportation networks present complex optimization challenges spanning calibration of high-fidelity simulators and real-time operational control. This paper presents a shared latent-space framework that connects simulator calibration and reinforcement learning control through a common learned representation of urban traffic dynamics. First, we develop a combinatorial MLP-autoencoder architecture that learns low-dimensional manifolds...

    arxiv.org/abs/2609.21945 · PDF

  12. 12

    Joint Remaining Useful Life Prediction and Capacity Estimation of Lithium-Ion Batteries Using Partial-Charging Data

    Khoa Tran, Ho-Si-Hung Nguyen, Phone Wai Yan Moe, Hung-Cuong Trinh, Thi-Hoang-Giang Tran

    cs.LG

    Joint remaining useful life (RUL) prediction and capacity estimation require representations of both gradual degradation and recent battery behavior. This paper presents a cross-expert framework using partial-charging measurements without measured historical full-cycle capacity as an input. The RUL Expert encodes nominal 10-min segments from ten cycles sampled within a 30-cycle history using a pretrained gated recurrent unit (GRU) encoder, a...

    arxiv.org/abs/2609.21932 · PDF

  13. 13

    Kinks vs. Smoothness: Identifiability of Real Analytic nICA for Laplace-like Sources

    Isaac Manring, Kejun Huang

    cs.LG

    Many machine learning systems try to explain complex data - like images or financial time series - in terms of hidden, independent factors that generated them. Recovering the true underlying factors, rather than some scrambled version of them, is the central challenge of nonlinear Independent Component Analysis (nICA). We prove identifiability (exact recovery) up to trivial ambiguities for real analytic generating functions when source...

    arxiv.org/abs/2609.21926 · PDF

  14. 14

    Beyond Kinematics: Benchmarking Simulation Fidelity for Muscle-Driven Imitation Learning

    Ayah G. Ahmad, Claire E. Borden, Maegan Tucker

    cs.LG · cs.RO

    In this work, we conduct a systematic comparison of two state-of-the-art motion-imitation reinforcement learning (MIRL) pipelines, one built on SCONE/HyFyDy and one built on MuJoCo/MyoSim. HyFyDy emphasizes physiological realism through detailed musculotendon modeling, while MuJoCo prioritizes computational efficiency and scalable policy learning. While recent work has demonstrated that both pipelines reproduce human kinematics with high...

    arxiv.org/abs/2609.21909 · PDF

  15. 15

    Intervention Granularity Matters: Coherent Treatment Bundles in Counterfactual Simulation with Clinical World Models

    Fangzhou Wang, Yixuan Yang, Camilla Balzarotti, Rishikesan Kamaleswaran

    cs.LG

    Counterfactual simulation with a clinical world model means fixing a patient's history, changing the treatment, and reading off the predicted response. Doing so requires deciding what counts as one intervention. In clinical settings, interventions are documented as bundles: a co-occurrence audit of 945,707 patient-hours from MIMIC-IV shows groups of components, such as every parameter of a dialysis circuit, that never appear apart, so an edit...

    arxiv.org/abs/2609.21906 · PDF

  16. 16

    ExpBoN: Exponential-Noise Best-of-$n$ for Efficient Test-Time LLM Alignment

    Yanxiao Liu, Sicheng Wan, Deniz Gündüz

    cs.LG · cs.IT

    Best-of-$n$ (BoN) sampling is a simple yet effective inference-time alignment method, but hard maximization provides only coarse control over the trade-off between reward and distribution shift. Soft Best-of-$n$ (Verdun et al. 2025) provides smoother control and converges to the optimal distribution associated with KL-regularized reward maximization. In this paper, we introduce ExpBoN, an alternative soft BoN method based on the...

    arxiv.org/abs/2609.21899 · PDF

  17. 17

    LLMs as Feature Engineers for Text-and-Tabular Prediction

    Merwan Barlier, Blaz Skrlj

    cs.LG

    We introduce an iterative framework that automates the extraction of interpretable, schema-bound categorical features from unstructured text for tabular prediction models. To navigate the feature space, a generator LLM proposes semantic definitions, a separate extractor LLM materializes the features, and a downstream tabular model evaluates their predictive performance. We optimize this search by translating explicit model errors, such as AUC...

    arxiv.org/abs/2609.21894 · PDF

  18. 18

    Detecting Pretraining Data in Large Language Models from a Free-Energy Perspective

    Chenye Ke, Zirui Liu, Qi Liu, Yan Zhuang, Jintao Zhang, Zhenya Huang, Shijin Wang

    cs.LG · cs.AI · cs.CL

    Detecting pretraining data in large language models is challenging because high likelihood can reflect either training exposure or strong generalization. In the joint space of prediction loss and predictive entropy, a likelihood-only detector uses a horizontal boundary and can mistake predictable non-members for members. Motivated by this, we introduce an inclined boundary that evaluates prediction loss relative to predictive entropy. Our...

    arxiv.org/abs/2609.21888 · PDF

  19. 19

    Geometric Mean Pooling for Equal-Weight Multiplicative Coarse-Graining

    Ang-Kun Wu, Fangdi Wen, Jingtao Zhang

    cs.LG · stat.ML

    As an alternative to the additive and extremal biases of average and max pooling, we introduce Geometric Mean Pooling (GMP), a signed pooling operator that combines the product of feature signs with the geometric mean of feature magnitudes. Motivated by local-to-global composition in quantum many-body physics, GMP retains both joint sign information and a characteristic multiplicative scale without introducing learnable pooling parameters. We...

    arxiv.org/abs/2609.21876 · PDF

  20. 20

    Neural Cellular Automata Learn General Features in their Hidden Channels

    Etienne Guichard, Stefano Nichele

    cs.LG · cs.AI

    Modern deep learning models achieve impressive generalization through over-parameterization, but this paradigm often struggles with overfitting and memorization in few-shot regimes. Neural Cellular Automata (NCAs) offer a highly parameter-efficient alternative, yet research has focused primarily on their output, leaving the role of their internal hidden channels largely unexplored. In this paper, we investigate the internal dynamics of NCA...

    arxiv.org/abs/2609.21870 · PDF

  21. 21

    The Weight Is Over - Interactive Diffusion on Consumer GPUs

    Frieder Ganz, Maximilian Müller

    cs.LG · cs.CV · cs.PF

    On-device inference is booming, but the momentum is almost all in language models. Diffusion pipelines are memory hungry, latency-sensitive, and require orchestrating an embedder, a transformer, a decoder, and often further postprocessing that is not as standardized as LLM inference loops are. We navigate the trade-off between performance, quality, and model footprint to reach as many client devices in the wild as possible. We make three...

    arxiv.org/abs/2609.21849 · PDF

  22. 22

    Federated Deep Clustering Networks for High-Dimensional and Heterogeneous Data

    Morris Stallmann, Charalampos S. Kouzinopoulos, Marcin Pietrasik, Anna Wilbik

    cs.LG · cs.AI

    Clustering high-dimensional data is a fundamental task in unsupervised machine learning with applications to a variety of domains. In the centralized data scenario, this task is commonly solved using deep clustering methods that utilize deep neural network architectures to learn clustering-friendly latent space representations. In Federated Learning, where data is distributed between clients and is private, deep clustering methods are less...

    arxiv.org/abs/2609.21829 · PDF

  23. 23

    Matrix AdaGrad: Row-wise and Column-wise Adaptive Subgradient Methods

    Wenpeng Zhang, Runsheng Yu, Peilin Zhao

    cs.LG · cs.AI

    Adaptive optimization methods such as AdaGrad and Adam are widely used in modern neural-network training, but their adaptive scaling is primarily designed for vector-valued parameters and does not explicitly exploit matrix structure. Recent matrix-aware optimizers demonstrate the benefits of structured optimization, yet a general theoretical framework for deriving matrix-aware adaptivity comparable to that of AdaGrad remains lacking. In this...

    arxiv.org/abs/2609.21815 · PDF

  24. 24

    RegKT: Interpretable and Robust Deep Knowledge Tracing With IRT-Regularizer

    Samuel Girard, Juan D. Pinto, Jill-Jênn Vie, Amel Bouzeghoub

    cs.LG

    As deep learning models continue to advance, knowledge tracing models have achieved higher accuracy. However, these gains come at the cost of reduced interpretability, which is crucial for practitioners in educational settings to adopt new methodologies. Additionally, deep learning models are prone to overfitting, particularly when dealing with the small datasets that are common in educational applications. In this paper, we propose a novel...

    arxiv.org/abs/2609.21791 · PDF

  25. 25

    Bilevel Optimization of Topology and Hyperparameters (BOTH)

    Suryanarayanan Manoj Sanu, Miguel Anibal Bessa, Alejandro Marcos Aragón

    cs.LG

    Topology optimization (TO) represents a significant step towards automating the design process: given a working simulation, TO can produce a viable prototype at the press of a button by differentiating the simulation and iteratively improving the design. In practice, however, TO is riddled with ``magic numbers''---hyperparameters whose tuning significantly affects the outcome. Finding the right values typically requires not only deep...

    arxiv.org/abs/2609.21758 · PDF

  26. 26

    GraphSkillEvo: Evolutionary Optimization of Graph-Structured Agent Skills

    Rui Sun, Zhi Zheng, Zhenkun Wang, Zhichao Lu

    cs.LG

    Skills can improve the performance of Large Language Model (LLM) agents by providing task-specific procedural guidance, while skill optimization further improves their effectiveness through iterative refinement. However, existing skill optimization methods typically represent skills as unstructured natural-language instructions, creating two key challenges: 1) Unstructured skills often lack explicit workflow-level guidance and contain...

    arxiv.org/abs/2609.21749 · PDF

  27. 27

    GEM-MPC: Balancing Exploration and Exploitation through Expert-Guided Planning

    Alvaro Serra-Gomez, Thomas Moerland

    cs.LG

    Effective exploration in high-dimensional continuous control remains a central challenge in reinforcement learning. Planning-based methods address this by combining online planning with learned policies and value functions, but their components can become misaligned during training: learned sampling policies may diverge from planner behavior, while planning distributions stored in replay become stale as the model and value function evolve....

    arxiv.org/abs/2609.21735 · PDF

  28. 28

    SpecQuant: Speculative Decoding with Multi-Parent Quantization for Adaptive LLM Inference

    Harish KB, Jagadeeswaran M, Pradheep P, Yuvanesh S, Sivakumar T

    cs.LG

    Running large language models (LLMs) locally continues to be limited by restrictions of compute and memory on consumer hardware. The popular acceleration technologies, such as quantization, speculative decoding, and adaptive inferencing, offer substantial speed boosts but usually necessitate retraining, per architecture tuning, or draft models. SpecQuant is a trainingfree framework, that combines speculative decoding with multiparent...

    arxiv.org/abs/2609.21704 · PDF

  29. 29

    Optimization Geometry of Equivalent Brownian RKHS Representations

    Mahdi Mohammadigohari, Gustau Camps-Valls

    cs.LG

    Equivalent finite parameterizations can represent the same functions and intrinsic norm yet induce different optimization algorithms. We study this effect in a controlled finite Brownian RKHS with nodal, increment, and spectral coordinates. Classical finite-element, RKHS-interpolation, Brownian-covariance, and mixed-boundary DCT identities make the shared hypothesis class, Brownian energy, approximation operator, and coordinate maps explicit....

    arxiv.org/abs/2609.21693 · PDF

  30. 30

    Multi-Domain Clustering via Measure Quantization

    Rafael Pereira Eufrazio, Eduardo Fernandes Montesuma, Charles Casimiro Cavalcante

    cs.LG · stat.ML

    Clustering is a fundamental task in data analysis, typically addressed through centroid-based methods such as K-means. In this work, we present a general framework for multi-domain clustering via measure quantization: given samples from multiple domains, we learn a shared set of cluster prototypes by minimizing a probability metric, such as the Sinkhorn divergence or the Maximum Mean Discrepancy, between each domain's probability measure and...

    arxiv.org/abs/2609.21664 · PDF

  31. 31

    Beyond Gaussian Worlds: Latent Geometry Matters for JEPAs

    Léo Nicollier, Enric Meinhardt-Llopis, Marc Pic, Pablo Musé, Gabriele Facciolo

    cs.LG

    Recent Joint-Embedding Predictive Architectures (JEPAs) prevent representation collapse by constraining learned representations to follow a prescribed target distribution, such as an isotropic Gaussian or the uniform distribution on a hypersphere. Klindt et al. (2026) showed that, under their Euclidean assumptions, matching a Gaussian target can recover Gaussian latent variables up to a linear transformation, and that the Gaussian is the...

    arxiv.org/abs/2609.21656 · PDF

  32. 32

    Riemannian Neural Hamiltonian Flows: Geodesic Symplectic Transport and Interpretability

    Vincent Souveton

    cs.LG

    Hamiltonian normalizing flows are attractive generative models because their phase-space maps are invertible and volume preserving, but most neural constructions are formulated in Euclidean space. We introduce Riemannian Neural Hamiltonian Flows, which combine the fixed kinetic energy of a Riemannian manifold, a learned scalar potential, and an explicit geodesic leapfrog integrator. Our analysis explains how the learned Hamiltonian can be...

    arxiv.org/abs/2609.21647 · PDF

  33. 33

    Trading Depth for Time in Recurrent Transformers

    Zeyi Huang, Xuehai He, Yong Jae Lee, Yelong Shen

    cs.LG

    Recurrent Transformers increase computational depth through temporal recurrence, feeding each token's high-level hidden state into the computation of the next. This raises a natural question: is additional computation better spent on more temporal steps or greater physical depth? We investigate this question using Latent Recurrent Transformers (LRTs), which retain one backbone forward pass per vocabulary token during decoding and provide a...

    arxiv.org/abs/2609.21605 · PDF

  34. 34

    On Repulsive and Attractive Teachers: Separating Correctness from Behavior in Self-Distillation

    Anton Baumann, Akmal Ashirmatov, Leo Schmidt-Traub, Frederike Lübeck, Jonas Hübotter, Thomas Kleine Buening, Andreas Krause

    cs.LG · cs.AI

    On-policy self-distillation provides dense, token-level supervision by conditioning a model on privileged information and distilling the resulting teacher distribution back into the model. However, privileged information can change not only what the teacher knows, but also how it behaves, entangling correctness-relevant learning signals with unintended behavioral shifts. We study this effect in reasoning tasks by contrasting attractive...

    arxiv.org/abs/2609.21561 · PDF

  35. 35

    OneBid: A Unified Auto-Bidding Foundation Model for Diverse oCPX Advertising Scenarios

    Yewen Li, Peng Jiang, Yitian Li, Pengfei Lv, Xialong Liu, Peng Jiang, Qingpeng Cai

    cs.LG · cs.AI

    Auto-bidding is central to computational advertising, where strategies must maximize advertisers' conversion value under economic constraints. It has evolved from rule-based controllers to reinforcement learning and generative methods such as Decision Transformer (DT). Yet these methods increasingly mismatch the prevailing optimized cost-per-X (oCPX) paradigm, which spans heterogeneous scenarios (e.g., registration, purchase), each served by...

    arxiv.org/abs/2609.21550 · PDF

  36. 36

    MACE: Memory-Agent Co-Evolution with Adaptive Memory Graphs for Multi-Agent Systems

    Kairui Yang, Minghao An, Xunkai Li, Ziheng Yi, Zekai Chen, Guangyuan He, Rong-Hua Li

    cs.LG

    LLM-based multi-agent systems generate collaboration traces that record how agents plan tasks, verify intermediate results, and repair failures. Reusing these procedures requires preserving an action's prerequisites and the outputs needed by subsequent agents. Our empirical studies show that grouping these dependencies into functional memory units improves their retention, while connecting units increases retrieval of the units and links...

    arxiv.org/abs/2609.21533 · PDF

  37. 37

    OpenMAS-GCom. A Diagnostic Benchmark for Graph-enhanced Multi-Agent Systems

    Kairui Yang, Xunkai Li, Kaixiang Zhang, Minghao An, Zekai Chen, Yuxuan Ba, Rong-Hua Li

    cs.LG

    Graph-enhanced multi-agent systems (G-MAS) coordinate large language model agents through communication graphs and role assignments, which determine how agents exchange information and divide responsibilities. However, final-score comparisons across systems combine differences in models, communication patterns, roles, and computation costs, making performance differences difficult to attribute to specific communication structures, role...

    arxiv.org/abs/2609.21527 · PDF

  38. 38

    IncentRL: The Trade-Off Between Preference Guidance and Task Performance

    Xuening Wu, Yanlan Kang, Shenqin Yin

    cs.LG

    Preference-based reward shaping can guide reinforcement learning, but adding preference signals to the reward may unintentionally change the task being optimized. We address this problem with IncentRL, a framework that introduces preference guidance while explicitly characterizing its effect on external-task performance. IncentRL adds a Kullback--Leibler (KL) penalty between a specified outcome distribution and a preferred distribution. For...

    arxiv.org/abs/2609.21525 · PDF

  39. 39

    What Must Survive? Exact Task-Information--State Frontiers for Resource-Sufficient Learning

    Ronald Katende

    cs.LG · cs.IT

    A system may be compressed before its downstream task is fully known. We ask how much retained state is then necessary and how much can be saved by limited advance task information. For a finite family of linear tasks, a task message is revealed before state formation and the exact task only afterwards. For an advice alphabet of size $K$, the exact frontier is \[ p^*(K)= \min_{\substack{\Pcal\text{ partition of }\U\\|\Pcal|\le K}}...

    arxiv.org/abs/2609.21523 · PDF

  40. 40

    Efficient Architecture Search under Leave-One-Subject-Out Evaluation

    Heinke Hihn, Friedhelm Schwenker

    cs.LG

    Deep neural architectures are widely used for signal processing in automated pain assessment systems. However, architecture design has remained largely a manual task despite the potential efficiency benefits of Neural Architecture Search (NAS). Embedding NAS in a Leave-One-Subject-Out (LOSO) evaluation is computationally demanding because a fully nested implementation requires $N$ independent architecture searches and, assuming approximately...

    arxiv.org/abs/2609.21457 · PDF

  41. 41

    Understanding LLM Quantization through Activation-Guided Compensation and Orthogonal Residuals

    Yamato Narita, Issei Sato

    cs.LG

    Post-training weight-activation quantization reduces the memory and inference costs of large language models, but aggressive W4A4 quantization remains difficult because activation outliers degrade effective quantization resolution. Although weight optimization, channel-wise scaling, and orthogonal rotation mitigate this problem, the error components they address and their relationship remain unclear. Using an exact decomposition of local...

    arxiv.org/abs/2609.21450 · PDF

  42. 42

    Optimal Randomized Proper Online Learning

    Zachary Chase, Idan Mehalel

    cs.LG

    We prove that the optimal expected mistake bound of online learning a function class $\mathcal{H}$ by a randomized proper learning algorithm is $O(\mathtt{L}(\mathcal{H}) \log T)$, where $\mathtt{L}(\mathcal{H})$ is the Littlestone dimension of $\mathcal{H}$ and $T$ is the time horizon. Our result improves upon the previously best known bound of $O(\mathtt{L}(\mathcal{H}) \log^6 T)$ given by Daskalakis and Golowich (STOC 2022), and is optimal...

    arxiv.org/abs/2609.21445 · PDF

  43. 43

    Decision-Focused Learning for Mean-Variance Portfolio Optimization via KKT-Based Reformulation

    Kensei Nosaka, Shunnosuke Ikeda, Yuichi Takano

    cs.LG

    Mean-variance portfolio optimization (MVO) is a central framework in data-driven asset management. A widely adopted approach is a two-stage framework that first predicts expected returns and then solves the optimization problem based on these predictions, with the predictive models trained by minimizing prediction errors. However, this objective of prediction is not aligned with the quality of the downstream portfolio decision....

    arxiv.org/abs/2609.21427 · PDF

  44. 44

    Tracing the Evidence Behind Zero-Shot Time-Series Forecasting: A Source-First Taxonomy and Audit Framework

    Delun Kong, Wanyun Ling, Chenxi Liu, Ziyue Li

    cs.LG

    Zero-shot time-series forecasting (TSF) is often described as forecasting without target-specific parameter updates, but that training-status condition does not specify what evidence the system may use. A frozen language model prompted with serialized values, a time-series model pretrained on broad forecasting corpora, and a retrieval-augmented forecaster may all satisfy the no-update condition while drawing on different transferable...

    arxiv.org/abs/2609.21425 · PDF

  45. 45

    Probabilistic Forecasting of Business Process Executions with Neural Temporal Point Processes

    Jiaxin Yuan, Daniela Grigori, Han van der Aa

    cs.LG

    Operators of service-based systems act on forecasts of how a running execution will continue, and such a forecast is actionable only if its reliability is known. Mainstream deep-learning models for this task are discriminative and deterministic: they emit a single next activity and a single remaining-time estimate, without a distribution to reason over. We instead cast the problem as generative sequence modelling with marked temporal point...

    arxiv.org/abs/2609.21382 · PDF

  46. 46

    Knowledge-Graph-Augmented Chronos-2 for HEC-RAS Surrogate Forecasting

    Edward Holmberg, Elias Ioup, Mahdi Abdelguerfi

    cs.LG · cs.AI

    We investigate whether coupling a time-series foundation model to hydraulic project knowledge improves surrogate forecasting of HEC-RAS water-surface elevation (WSE). We present KG-Chronos-2, which combines a frozen Chronos-2 predictor with exact-state residual decoding, graph-conditioned historical retrieval, and input-aligned correction. We compare the method with persistence, a residual LSTM, project-conditioned recurrent GeoFNO, a...

    arxiv.org/abs/2609.21381 · PDF

  47. 47

    IntBMoE: Integrating Block-Level Conditioning into Expert Composition for Full-Participation Mixture-of-Experts

    Ran Cheng, Longfei Xu, Zheng Liu, Kaikui Liu, Xiangxiang Chu

    cs.LG

    Mixture-of-Experts (MoE) scales capacity, but existing designs cannot set three quantities independently. For a single token, participation is how many experts contribute knowledge to its output, execution is how many are actually computed (compute cost), and materialization is how many expert-sized parameter sets must be built and stored (memory cost). Sparse routing keeps execution and materialization low, but shrinks participation: for...

    arxiv.org/abs/2609.21346 · PDF

  48. 48

    Routine Blood Tests Outperform CRP for Distinguishing Bacterial From Viral Infection in Children

    Mihaela Demireva, Zhecho Mitev, Djuna Chinareva-Klimentova, Svetoslav Ivanov, Georgi Nalbantov, Dimitar Mitev

    cs.LG

    Acute infectious diseases are among the leading causes of medical consultations and hospitalizations in children worldwide. These infections are predominantly caused by viruses or bacteria, yet differentiating between the two remains a common clinical challenge. As a result, pediatricians often default to the safer option of prescribing antibiotics contributing to the growing problem of antimicrobial resistance. The objective is to assess the...

    arxiv.org/abs/2609.21332 · PDF

  49. 49

    Deep Reinforcement Learning with Buffered Quantile Objectives

    Mohammad Alipour-vaezi, Sajad Khodadadian

    cs.LG · cs.AI · math.OC

    Quantile-based reinforcement learning provides an interpretable approach to risk-sensitive decision-making by optimizing a prescribed quantile of the cumulative-return distribution. Despite this appeal, learning under a point quantile objective is challenging: quantiles can change abruptly under small perturbations of the return distribution, and exact quantile-sensitive planning requires computationally demanding distributional optimization....

    arxiv.org/abs/2609.21327 · PDF

  50. 50

    An Introduction to Compression-Based Machine Learning

    John Hurwitz, Edward Raff, Charles K. Nicholas

    cs.LG

    Any lossless compression algorithm (like gzip) may be converted into a machine learning method, via either Normalized Compression Distance or the Minimum Description Length principle. Any auto-regressive model may be converted into a lossless compression method via entropy coding. This seemingly circular dependence has unrealized potential in modern artificial intelligence and machine learning, and we survey and formalize the various...

    arxiv.org/abs/2609.21309 · PDF

  51. 51

    Fast And Accurate Text Content File Type Identification

    Manu Nandan, Michael Brautbar, Edward Raff

    cs.LG

    A common requirement across organizations is to have a tool that can identify file types based on their contents, particularly in the cybersecurity domain where magic numbers and file extensions can not be trusted. While existing tools work well in practice, there is plenty of room for improvement either in terms of computational load and time for detection in the case of model based tools like Magika or in terms of accuracy of detection in...

    arxiv.org/abs/2609.21306 · PDF

  52. 52

    FairLMs: A Turnkey Library for Fairness in Language Models

    Jiale Zhang, Michael Larionov, Zichong Wang, Zhipeng Yin, Wenbin Zhang

    cs.LG · cs.CL

    Fairness research on language models involves measuring bias, applying mitigation methods, and examining the evidence on which an evaluation rests. Existing tools offer complementary functionality through different interfaces, so combining them requires reconciling model interfaces, evidence formats, access constraints, and result types before applicability can be checked or methods compared. We introduce \textbf{FairLMs}, a Python library...

    arxiv.org/abs/2609.21296 · PDF

  53. 53

    Multi-Subject Pretraining Enables Short-Calibration Personalization for Closed-Corpus Surface EMG Speech Decoding

    Chenqian Le, Beatrice Fumagalli, Yasamin Esmaeili, Xupeng Chen, Tianyu He, Nikasadat Emami, Adeen Flinker, Yao Wang

    cs.LG

    Surface electromyography (sEMG)-based silent speech interfaces are limited by cross-user variability and calibration burden. We study a limited-data setting in which each of 27 speech-typical participants contributed less than 0.5 h of data (21.3 min on average) across Aloud and Mimed speech. Within a closed 50-sentence corpus, we used leave-one-subject-out evaluation, initializing from a released single-subject checkpoint, pretraining on...

    arxiv.org/abs/2609.21288 · PDF

  54. 54

    MIRCID: Inferred Hub-miRNAs Drive Cross-Task Improvements in Drug Mechanistic Modeling

    Xin Cao, Yigang Chen, Jiatong Xu, Ziyue Zhang, Xiang Cheng, Shenyu Wang, Yangyi Zhang, Xiaoxuan Cai, Shidong Cui,...

    cs.LG

    Drug mechanism-of-action (MoA) modeling commonly relies on perturbational transcriptomes, but matched microRNA (miRNA) measurements are often unavailable. Inferred regulatory features offer a scalable way to reuse these data. Here, we present MIRCID, a framework comparing gene expression with inferred transcription factor (TF) activity and miRNA expression across pathway classification and similarity-based MoA retrieval. HubmiRNet infers 414...

    arxiv.org/abs/2609.21280 · PDF

  55. 55

    Reliability-Centered Evaluation of Sparse Longitudinal CT Lesion-Size Forecasting with Conformal Interval Calibration and Gompertz-Inspired Regularization

    Lingfei Kong

    cs.LG · q-bio.QM · stat.ME

    Sparse longitudinal CT follow-up limits lesion-size forecasting when only a few prior observations are available. We constructed a five-visit DLT-derived same-lesion trajectory benchmark from DeepLesion and Deep Lesion Tracker (DLT), yielding 205 trajectories from 129 patients. We compared an exploratory conventional sparse-to-final analysis with a primary fixed visit-index horizon design predicting the common log change from T3 to T4 while...

    arxiv.org/abs/2609.21197 · PDF

  56. 56

    SWE-Proof: Can Language Models Resolve Real-World Issues with Machine-Checked Proofs?

    George Ma, Benjamin Mikek, Haoyu Li, Ferhat Erata, Yuhao Zhang, Zeren Shui, Behrooz Omidvar Tehrani, Jun Huan,...

    cs.LG · cs.AI · cs.SE

    Ensuring the correctness of LLM-generated code is a core challenge for modern software engineering. Benchmarks for agentic code generation check correctness with held-out test suites, which are inherently incomplete and increasingly susceptible to memorization. Formal verification avoids both problems, but existing work covers only standalone tasks whose specifications are given as input, not real issues, which touch large repositories and...

    arxiv.org/abs/2609.21190 · PDF

  57. 57

    TierKV: Long-Context On-Device LLMs via Predictive Multi-Tier KV Caching

    Zhihao Shu, Md Musfiqur Rahman Sanim, Jie Hu, Kun Yuan, Minghai Qin, Gagan Agrawal, Wei Niu

    cs.LG · eess.SY

    Large language models (LLMs) are moving onto mobile devices for increasingly diverse workloads over text, images, video, and audio. These applications often require long contexts, making the Key-Value (KV) cache a dominant memory bottleneck because it grows linearly with sequence length and is accessed at every decoding step. Prior work reduces KV-cache footprint through low-rank compression, token eviction, or flash offloading, but the...

    arxiv.org/abs/2609.21172 · PDF

  58. 58

    M2G-LLM: Enhancing Clinical Prediction via Multimodal Graph Reasoning and LLM Context Injection

    Inyoung Choi, Sukwon Yun, Jiayi Xin, Jie Peng, Tianlong Chen, Qi Long

    cs.LG

    Integrating diverse data modalities --- such as clinical notes, laboratory results, and medical imaging --- is essential for advancing clinical decision-making. While Large Language Models (LLMs) have shown remarkable performance in processing unstructured clinical text, their limited capacity to incorporate non-text modalities hinders their broader utility in healthcare applications. Here, we introduce M2G-LLM (Multimodal MedGraph-LLM), a...

    arxiv.org/abs/2609.21164 · PDF

  59. 59

    HMB-GAN: Hybrid Multi-Bézier GAN for Vector Shape Synthesis

    Elian Hugh Thiele-Evans, Binh Duong Pham, Hani Omar M Alharbi, Liibaan Aaden, Syed Umer Hasnain Zaidi, Prem Prakash...

    cs.LG

    We explore the use of hybrid quantum-classical generative adversarial networks for synthesising CAD-ready vector geometries. Unlike prior work that operates in rasterised or single-Bézier domains, we introduce HMB-GAN (Hybrid Multi-Bézier GAN), an end-to-end differentiable generative framework that constructs closed shapes through stitched multi-segment Bézier representations with geometric continuity enforced by construction. We compare a...

    arxiv.org/abs/2609.21158 · PDF

  60. 60

    EnSol: an environment-aware graph neural network for molecular solubility prediction

    Thao Nguyen, Saman Shafaei, Zhengyi Zhang, Huimin Zhao, Heng Ji

    cs.LG · cs.AI

    Molecular solubility directly affects key aspects of molecular development such as reaction feasibility, formulation performance, separation efficiency, and solvent selection. However, experimental measurement across solutes, solvents, and temperatures remains costly and sparsely sampled. Existing computational models often rely on fixed-solvent assumptions, deterministic formulations, or simplified representations of solute-solvent...

    arxiv.org/abs/2609.21151 · 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.