cs.LG · 2026-11-11 · No. 170

Machine Learning, 2026-11-11.

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

01 — The papers

68 entries
  1. 01

    TorchCurves: Differentiable Parametric Curves

    Alex Shtoff

    cs.LG · stat.ML

    We present TorchCurves, a PyTorch library that treats vector-valued parametric curves as learnable primitives. Its central operation maps batches of scalar inputs to points on curves represented in B-spline or Legendre polynomial bases, with differentiation through both inputs and coefficients. Curve coefficients can be optimized jointly with the surrounding model or produced by another network. The library provides trainable modules,...

    arxiv.org/abs/2611.14516 · PDF

  2. 02

    AutoScaffold: AI-Guided Scaffolding for Reinforcement Learning of Language Models

    Zijian Zhang, Rizhen Hu, Jiangweizhi Peng, Shiyang Li, Sujay Bhatt, Sihan Zeng, Mingyi Hong

    cs.LG · cs.CL

    Reinforcement learning with verifiable rewards (RLVR) produces rich training experience, yet outcome rewards reduce each attempt to a single number and leave most of this information unused. Training-time scaffolds, such as strategies or hints added to the prompt during training, offer a natural way to turn this information into guidance, but their design and adaptation are largely left to humans and tailored to specific domains. We introduce...

    arxiv.org/abs/2611.14515 · PDF

  3. 03

    Pyrite: Building and Benchmarking Protein-Ligand Pose Ranking Agents

    Megha Srivastava, Michael H. Lim, Allison M. Keys, Pranav Murugan, Nate Gruver

    cs.LG · q-bio.QM

    Predicting how a ligand binds to its target is central to small-molecule drug discovery. Modern cofolding models can generate diverse candidate poses, but reliably ranking these candidates remains challenging. We introduce Pyrite, a pairwise pose-ranking benchmark comprising 1,500 pose pairs across diverse protein-ligand complexes. Expert chemists equipped with 3D visualization tools and physics-based scores perform near chance, highlighting...

    arxiv.org/abs/2611.14511 · PDF

  4. 04

    When Does the Real Wiring Win? Testing connectome structure-function claims against unmeasured physiology

    Yeyuan Liu

    cs.LG · q-bio.NC

    Connectome-constrained models are often used to argue that a measured wiring diagram outperforms shuffled or rewired wiring, and hence that structure shapes function. Such comparisons can confound wiring with how the models are tuned: a connectome fixes who connects to whom, not the excitability, gains or plasticity rates the model also needs, and modellers often tune these until the real-wiring model works. We separate three estimands:...

    arxiv.org/abs/2611.14508 · PDF

  5. 05

    Continual-ARC: Measuring Skill Acquisition over a Working Lifetime

    Daniel Lenton

    cs.LG

    Agent harnesses have emerged as a promising mechanism for continual learning, alongside test-time parametric learning. However, the current benchmarks have several limitations: they generally focus on a single applied domain, serve each instance only once, and assume a single family of learner, so memory-based and weight-based systems are rarely compared on the same scale. We propose Continual-ARC, a lifelong stream built from the three...

    arxiv.org/abs/2611.14505 · PDF

  6. 06

    Emergent Biological Capabilities in a Foundation Model for Molecular Interactions

    Adi Mashiach

    cs.LG · q-bio.BM

    Foundation models trained on large unlabeled corpora develop emergent capabilities, tasks that go beyond the training objective. We extend this paradigm to multimodal biology with Bonbon, a foundation model for molecular interactions trained on approximately 3.2 trillion tokens of paired protein-ligand sequences. From a single representation space, Bonbon exhibits three zero-shot emergent capabilities at hierarchical levels of resolution:...

    arxiv.org/abs/2611.14504 · PDF

  7. 07

    TimeSeek: Multi-Task and Multi-Domain Pretraining for Generalized Time Series Analytics

    Peng Chen, Yihang Wang, Zhongwen Rao, Yang Shu, Yuying Qiu, Yuxuan Chen, Lujia Pan, Bin Yang, Chenjuan Guo

    cs.LG

    Time series analytics has been widely applied across various domains. While existing time series foundation models (TSFMs) have demonstrated promising results, they still face two major challenges: 1) Most TSFMs mainly rely on stacking Transformer layers to increase parameter size, lacking the ability to adaptively model complex time series patterns from multi-domain data. 2) Current TSFMs are primarily pre-trained for specific tasks, e.g.,...

    arxiv.org/abs/2611.14491 · PDF

  8. 08

    A Theory of Policy Optimization in Post-Training: Coverage Collapse and Reference Resets

    Audrey Huang, Nan Jiang, Sham M. Kakade, Sadhika Malladi, Akshay Krishnamurthy, Dylan J. Foster

    cs.LG · stat.ML

    Policy optimization is central to language model post-training, yet principled guidance remains limited. We revisit its theoretical foundations to identify algorithms that learn near-optimal policies using total samples and gradient steps governed by the base model's response coverage, a measure of how readily it generates optimal responses. We call such algorithms coverage-preserving. This guarantee captures the goal of amplifying...

    arxiv.org/abs/2611.14467 · PDF

  9. 09

    Accurate predictions of properties of energetic materials with tabular foundation models

    Vagn Krister Nes Glomsrud, Mathias Bynke, Lars Henry Berge Olsen, Dennis Christensen

    cs.LG

    We use TabPFN, a recently developed tabular foundation model, to predict three properties of energetic materials: impact sensitivity, decomposition temperature and crystal density. Without any hyperparameter tuning, TabPFN outperforms the best published models for all three properties. It also substantially outperforms a wide range of established machine-learning models, including tuned gradient boosting (XGBoost), random forests and Gaussian...

    arxiv.org/abs/2611.14462 · PDF

  10. 10

    Quantifying the Sampling Frequency Invariance of Time-Series Foundation Models for Sensor Data Applications

    Sourish Gunesh Dhekane, Abinaya Kumar, Torsten Woertwein, Yan-Bo Lin, Thomas Ploetz, Shabnam Ghaffarzadegan

    cs.LG · cs.AI

    Time-Series Foundation Models (TSFMs) are increasingly used in sensor applications, where sampling frequency (SF) varies across sensors, devices, and operating modes (e.g., power saving). Ideally, TSFMs must remain effective during such SF changes, provided the SF retains task-critical bandwidth. In this paper, we stress-test 4 prominent TSFMs: MOMENT, Chronos-2, Moirai-MoE, and TimesFM, to quantify their SF-invariance and investigate whether...

    arxiv.org/abs/2611.14425 · PDF

  11. 11

    DnNP: Denoising Input Uncertainty in Neural Processes

    Fatemeh Tohidian, Chengzhi Shi, Soomi Lee, Matthew Elia, Amy Mueller, Stratis Ioannidis, Jennifer Dy

    cs.LG

    Neural Processes (NPs) are flexible meta-learning models capable of uncertainty quantification across diverse applications. However, the standard NP framework assumes access to noise-free inputs. This is unrealistic in many practical settings, where sensor noise or measurement errors are unavoidable. This discrepancy between model assumptions and real-world data can significantly degrade prediction quality and uncertainty estimates. We...

    arxiv.org/abs/2611.14424 · PDF

  12. 12

    Causal Concept Explanations for Deep Neural Models

    Josh Rountree, Pulkit Verma, Oswin So, Chuchu Fan, Julie A. Shah

    cs.LG · cs.AI

    Concept-based explanations in neural models are important for the development, certification, and adoption of autonomous systems. Yet, interpretability remains correlational. Without methods to guide and quantify causal relationships between explanations, a concept explanation can be provided without being necessary or sufficient for the observed output thereby appearing correct while misleading. We introduce Causal Concept Wrapper Network...

    arxiv.org/abs/2611.14407 · PDF

  13. 13

    Discovering Equations for the Thermodynamics of South Asian Monsoon Precipitation

    Savannah L. Ferretti, Jane W. Baldwin, Michael S. Pritchard, Tom Beucler

    cs.LG · physics.ao-ph

    South Asian monsoon precipitation depends on the thermodynamic structure of the lower troposphere, but how its ingredients combine to shape precipitation remains unclear, partly because this dependence is nonlocal in the vertical. We use symbolic distillation, which combines neural networks and symbolic regression, to discover interpretable precipitation equations from 21 years of June-August reanalysis data. Neural networks first learn the...

    arxiv.org/abs/2611.14401 · PDF

  14. 14

    REAL Loss: Optimizing Real Expected Acceptance Length for Speculative Decoding

    Yu Tian

    cs.LG · cs.CL

    Speculative decoding accelerates LLM inference by having a draft model propose multiple tokens for the target model to verify in one pass. The speedup is governed by the expected acceptance length (EAL), yet standard training objectives such as cross-entropy, KL divergence, and total variation focus only on distribution matching or single-step overlap. Recent studies instead propose block-level training objectives, aiming to optimize EAL...

    arxiv.org/abs/2611.14399 · PDF

  15. 15

    A Margin Perspective on LoRA: Robustness to Catastrophic Forgetting and Adapter Merging (MaLoRA)

    Ziqing Xu, Hancheng Min, Lachlan Ewen MacDonald, Salma Tarmoun, Enrique Mallada, Weijie Su, Rene Vidal

    cs.LG

    Low-rank adaptation (LoRA) is the de facto method for parameter-efficient fine-tuning of large neural networks, achieving strong task adaptation with minimal trainable parameters and low optimization cost. Beyond this efficiency, practitioners have observed two striking properties: \textit{robustness to catastrophic forgetting} and the ability to \textit{merge independently trained adapters} into a single model that performs competitively...

    arxiv.org/abs/2611.14397 · PDF

  16. 16

    Uncertainty-Aware Hybrid Offline Reinforcement Learning with Online Action Selection Enabled by Sparse Gaussian-Process Critics

    Marco Chan, Christos Evripidou, Liqiu Dong, Benoit Chachuat, Mehmet Mercangöz

    cs.LG · eess.SY

    Start-ups and grade changes remain largely operator-driven, while developing and maintaining dynamic models for model-based control requires substantial engineering effort. Hybrid Offline Learning and Online Optimisation (HOFLON) uses one-step offline reinforcement learning to select actions from historical transition data. In its original implementation, the online objective combines the predicted return from a tree-ensemble critic with an...

    arxiv.org/abs/2611.14374 · PDF

  17. 17

    The Unused Input: The Initial State as a Conditioning Interface for Recurrent TSFMs

    Bernhard Voggenberger, Christian Ganhör, Levente Zólyomi, Sebastian Böck, Sepp Hochreiter

    cs.LG

    Time series foundation models (TSFMs) forecast unseen series without task-specific training. Increasingly they are built on recurrent backbones, whose constant per-step cost makes long-context inference cheap, and which, unlike transformers, carry a state. A recurrent backbone computes its prediction from two inputs: the observed context window and the initial recurrent state. In practice only the first is utilised and the initial state is...

    arxiv.org/abs/2611.14357 · PDF

  18. 18

    Forward-Value Discretization: Faster Stochastic Diffusion Sampling Without High-Order Approximation

    Yuchen Jiao, Na Li, Changxiao Cai, Yuting Wei, Gen Li

    cs.LG · cs.AI · stat.ML

    Stochastic diffusion samplers often require substantially more neural function evaluations (NFEs) than deterministic samplers to achieve comparable sample quality. Although higher-order discretization accelerates deterministic sampling, its benefits for stochastic sampling remains unclear. We establish matching upper and worst-case lower bounds showing that stochastic samplers with higher-order approximations remain first-order convergent,...

    arxiv.org/abs/2611.14328 · PDF

  19. 19

    Auditing the Judge: Internal Signals for LLM Reasoning Assurance

    Mayank Gulati, Gerhard Wunder

    cs.LG

    Language-model systems use self-judgment to decide which answers to release, review, or recompute. Because intervention is costly, an assurance score must rank errors well at the available budget, not merely over the full ranking. We compare the judge's continuous verdict-token score with a supervised correctness probe trained on attention-head activations from the same prompt. Exact identities link fixed-budget error recall to residual risk...

    arxiv.org/abs/2611.14323 · PDF

  20. 20

    PSL-RGHA: Parallel Split Learning with Reference Gradient Half-Space Alignment for Mitigating Gradient Divergence

    Han Li, Zheng Lin, Miao Yang, Zehang Lin, Jing Yang, Zihan Fang, Qiyang Zhang, Zhaolu Kang, Jianmin Li, Lin Li,...

    cs.LG · cs.NI

    The escalating complexity of deep neural networks hinders the deployment of federated learning (FL) on resource-constrained devices, especially for multimedia applications such as on-device image classification and speech recognition. Parallel split learning (PSL) has emerged as a promising solution by offloading a substantial portion of the computational workload to the server through layer-wise model partitioning, making it particularly...

    arxiv.org/abs/2611.14312 · PDF

  21. 21

    MMDM: Measuring Massive Multitask Decision Making

    Junhao Shen, Qing Zong, Yue Guo, Zhanqiu Zhang, Yiwen Guo, Hong Cheng

    cs.LG · cs.CL

    System-1 Decision Models directly make structured probabilistic decisions over query-specified options given a state in a single forward pass, without autoregressively generating reasoning trajectories or textual responses. Recent evaluations cover diverse tasks but often rely on existing benchmarks, where prior exposure to test material can obscure generalization. Moreover, real-world decision making often involves long contexts, large...

    arxiv.org/abs/2611.14309 · PDF

  22. 22

    Federated Learning for Cross-Regional Epidemic Forecasting with Asynchronous Outbreaks

    Ruibin Liu, Francisco Fernandes, Xuanfeng Li, Zige Liu, André Salgado, Chitin Hon, Zifeng Yang, Arlindo Oliveira,...

    cs.LG

    Objective: Epidemic outbreaks are often asynchronous across regions, yet collaborative forecasting is hampered by privacy laws and heterogeneous public health interventions. We developed and evaluated, as a proof of concept, a federated learning (FL) framework for cross-regional COVID-19 forecasting under asynchronous outbreaks among Guangdong, Macau, and Portugal, while maintaining data sovereignty. Methods: Using daily case data from WHO...

    arxiv.org/abs/2611.14272 · PDF

  23. 23

    Physics-Constrained Autoencoder for Interpretable Health Representation Learning in Turbofan Engines

    Leonardo Baldo, Jokin Alcibar Ibarzabal, Daniel Cisneros Acevedo, Kai Goebel, Manuel Arias Chao

    cs.LG · eess.SP

    Health Indices (HIs) summarize degradation from operational data and support diagnostic and prognostic tasks in Prognostics and Health Management (PHM). Their usefulness depends on capturing degradation and interpreting system condition. Autoencoders learn such representations from unlabelled monitoring data; however, when training is driven only by reconstruction, the physical meaning of latent variables remains undefined, limiting their...

    arxiv.org/abs/2611.14266 · PDF

  24. 24

    Chain of Physics: Supporting Fault Diagnosis with Engineering-Inspired Large Language Model Reasoning

    Wenqi Zhou, Leonardo Baldo, Kai Goebel, Manuel Arias Chao

    cs.LG · eess.SP

    Fault diagnosis in complex systems requires linking observed changes in physical quantities to the failure mechanisms that produce them. Experienced engineers perform this task by combining sensor information with knowledge of system behavior and fault propagation. Large Language Models (LLMs) can support this process, but their diagnostic reasoning depends on the availability of asset-specific engineering knowledge and on how this knowledge...

    arxiv.org/abs/2611.14263 · PDF

  25. 25

    SinkState: On-the-Fly SVD of Recurrent States for Efficient Hybrid LLM Serving

    Haolin Jiang, Weihang Li, Zhongzhu Zhou, Jinwen Pan, Santosh Pandey, Hang Liu

    cs.LG

    Hybrid models combine softmax attention with linear attention, which compresses history into fixed-size recurrent states to balance model quality and inference efficiency. However, linear attention has not pushed this compression to its limit: we find that the queries, keys, and values of linear layers concentrate in a stable, low-dimensional subspace, a phenomenon we call the linear attention sink, so each state is effectively low-rank, with...

    arxiv.org/abs/2611.14246 · PDF

  26. 26

    Verifiable Agentic Traffic Class: A Policed Quality-of-Service Mechanism for Agentic AI Traffic, Evaluated from Simulation to Real-Kernel Adversarial Testing

    Naveen Babu Narayanan Nair, Sajay Kurup

    cs.LG · cs.AI · cs.NI

    Software agents built on large language models generate traffic unlike human web traffic, yet routers have no class for it and cannot verify a sender's claim to be an agent. This paper proposes the Verifiable Agentic Traffic Class (VATC), which combines a reserved Differentiated Services Code Point (DSCP) value declaring agent traffic, a router-side classifier that demotes flows contradicting the declaration to a class served after...

    arxiv.org/abs/2611.14211 · PDF

  27. 27

    Tree-Ising: Repairing Language Models with Bit-Level Ising Edits

    Hadi Salloum, Ammar Ali

    cs.LG

    Weight-space programming, introduced in recent MIT work by Takehi et al., enables a frontier LLM agent to read a network's weights and write a repair. Their GPT-6 Astra agent corrected the reasoning loops of Qwen3.5-0.8B without gradients, using 158 agent turns, 89.7 million tokens, and 50 forward-pass requests. Our method, Tree-Ising, casts this repair as a sparse discrete optimization problem over the model's own BF16 weights. Forward...

    arxiv.org/abs/2611.14210 · PDF

  28. 28

    Unipolar Neuron Networks Trained by Recirculation for Reward-Based Continual Learning

    Florin Leon

    cs.LG

    Unipolar neurons are the predominant type of nerve cell in the octopus and in most other invertebrates. A single neurite (a fiber extending from the cell body) branches into a tree that carries both input and output synapses. The branches send graded local outputs to nearby cells, and one site on the main neurite sends a global output to distant targets. This paper proposes a network of such neurons and a learning rule for it. Each neuron is...

    arxiv.org/abs/2611.14153 · PDF

  29. 29

    Deviation-Aware Pretraining for Cardiovascular Disease Detection from Electrocardiograms

    Siyuan Yan, Haibo Zheng, Feng Lu, Wei Li

    cs.LG

    Electrocardiogram (ECG) signals are critical for cardiovascular disease (CVD) detection. Most current deep learning models rely on supervised, task-specific training, which limits their ability to generalize across diseases, patient populations, and acquisition settings. This paper presents a self-supervised ECG feature extraction framework that learns from nearly one million unlabeled 12-lead ECG recordings. The framework includes two...

    arxiv.org/abs/2611.14141 · PDF

  30. 30

    HaiRuo: Forecasting Ocean Eddies as Fields and Objects with a Satellite-Altimetry Foundation Model

    Xukun Qin, Haochen Sun, Yuyang Tian, Ge Chen

    cs.LG · cs.CV

    Mesoscale eddies carry most of the ocean's kinetic energy and move heat, salt and nutrients across basins, yet where an eddy will be two weeks ahead is still read off ocean models with post-hoc tracking heuristics. We introduce HaiRuo, which forecasts eddies as fields and as objects at once: from 30 days of satellite sea-level maps, one forward pass returns 15 days of sea level and geostrophic velocity, daily eddy masks and the trajectory of...

    arxiv.org/abs/2611.14136 · PDF

  31. 31

    Right for the Wrong Reasons? Evaluating Plausibility and Behavioral Change in Explanation-Guided Learning

    David Dembinsky, Adriano Lucieri, Andreas Dengel

    cs.LG

    Explanation-Guided Learning (EGL) promises models that are not only accurate but also rely on task-relevant evidence. However, whether improved agreement between model explanations and human-provided rationales also implies a changed decision strategy remains unclear. We study this tension under controlled synthetic shortcut introduction using a constructed shortcut dataset. Inspired by real-world applications, our benchmark covers shortcuts...

    arxiv.org/abs/2611.14134 · PDF

  32. 32

    Hierarchical Graph Representation Learning with Pooling-Induced Substructures

    Luca Sbicego, Xiaowen Dong, Dorina Thanou

    cs.LG

    Hierarchical graph-level tasks that require reasoning over substructures that extend beyond local neighborhoods are posing challenges for standard Graph Neural Networks (GNNs). While graph pooling aims to address this by coarsening graphs into hierarchical representations, existing methods often fail to explicitly preserve the substructures they induce, leading to inconsistent empirical performance. Moreover, evaluating such approaches...

    arxiv.org/abs/2611.14126 · PDF

  33. 33

    Opportunity-Cost Option-Critic: Adaptive Switching Costs for Robust Hierarchical Reinforcement Learning

    Khaled Ammoura, Aya Kayal, Naseem Daher

    cs.LG

    Deliberation costs are commonly used in hierarchical reinforcement learning (HRL) to prevent options from collapsing to primitive actions. Existing approaches rely on fixed penalties or duration-based adaptations, both of which are indirect proxies for an option's utility. This paper argues that option persistence should instead depend on the opportunity cost of abandoning the current behavior. We propose the Opportunity-Cost Option-Critic...

    arxiv.org/abs/2611.14099 · PDF

  34. 34

    Generalizing Causal Decision-Making: Markov Equivalence, Transportability and Counterfactuals

    Min Woo Park, Sanghack Lee

    cs.LG

    Structural causal bandits use causal knowledge to guide exploration when reward mechanisms are unknown. This paper extends the approach to partially known causal structure, information from other environments, and realizable counterfactual actions. First, we develop graphical criteria for minimal intervention sets under Markov equivalence and characterize possibly-optimal minimal intervention sets for maximal ancestral graphs. A search...

    arxiv.org/abs/2611.14066 · PDF

  35. 35

    Structured Confidence Perturbation via Fisher Geometry for Adversarial Training

    Chenlong Wei, Hanling Wang, Chong Zhang, Qinkai Yu, Fangyu Wu, Xiaobo Jin

    cs.LG · cs.CV

    Adversarial training is often plagued by "pseudo-attacks", perturbations that significantly increase the loss value but fail to truly alter the model's confidence structure. To address this, we propose two complementary approaches to directly attack and defend against confidence structure deformation. First, building upon the traditional loss maximization perspective, we introduce a hybrid objective that integrates cross-entropy and KL...

    arxiv.org/abs/2611.14046 · PDF

  36. 36

    ScaleBrain: A Scale-Consistent Framework for EEG Foundation Models

    Minyu Zhou, Yajing Wu, Yongqiang Tang, Wang Jia, Wensheng Zhang

    cs.LG · eess.SP · q-bio.NC

    Foundation models pre-trained on large unlabeled electroencephalography (EEG) corpora have emerged as a transformative paradigm for Brain-Computer Interface (BCI) applications. Recent EEG foundation models have shown that multi-scale representations derived from fixed-length patches can improve EEG representation learning. Existing frameworks commonly construct temporal scales through feature-level operations. However, these operations leave...

    arxiv.org/abs/2611.14035 · PDF

  37. 37

    Margin-Guided Token Credit for Reinforcement Learning with Verifiable Rewards

    Chi Liu, Xin Chen, Yuchuan Weng, Srinivasan Manoharan

    cs.LG · cs.AI

    In reinforcement learning with verifiable rewards, a language model is rewarded for a correct final answer, and GRPO passes that reward to every token of the solution in equal measure. We instead place the credit where the model hesitated, in a method we call MCPO (Margin-Credit Policy Optimization). We measure hesitation by the top-2 margin $M=p_1-p_2$ of the policy's next-token distribution, map it to a score with two fixed thresholds, and...

    arxiv.org/abs/2611.14015 · PDF

  38. 38

    Coding Agent Memory Post-Training II: Unlocking the Memory Potential of Pre-trained File Operations for Long-Horizon Tasks via On-Policy Distillation

    Lirui Luo, Kelong Mao, Xinwei Yang, Rongqing Li, Luyu Chen, Kieran Wong, Yudong Guo, Xinrui Wang, Jiayin Zhu, Gu...

    cs.LG · cs.AI · cs.CL

    Language-model agents increasingly tackle long-horizon tasks whose interaction histories exceed the model's active context. They must therefore decide what to write down and what to read back, and when to do either. Recent work has begun to use on-policy distillation (OPD) to make memory control part of the policy, relying on memory tools and compaction protocols that lie outside either model's pre-training. However, this weakens the...

    arxiv.org/abs/2611.14009 · PDF

  39. 39

    BioXArena-Pro: a research-grade benchmark for autonomous biomedical machine learning agents

    Loka Li, Duzhen Zhang, Xingbo Du, Leonard Song, Kun Zhang, Le Song

    cs.LG

    Coding agents are increasingly proposed as autonomous builders of biomedical models, but existing benchmarks are small, near saturation or open to leaked information. We present BioXArena-Pro, 64 research-level modelling tasks from eight biomedical domains, each rebuilt from a published study; most hold out whole biological units such as species or cohorts. Across 22 agent configurations, the best scores 55.1 of 100, and the scaffold matters...

    arxiv.org/abs/2611.13994 · PDF

  40. 40

    Temporal Behavior Trees for Reinforcement Learning: Specifications as Rewards, Curricula, and Diagnostics

    Vladimir Krsmanovic, Bernd Finkbeiner

    cs.LG

    Reinforcement learning from formal specifications often separates task definition, reward shaping, curriculum design, and policy diagnostics into different mathematical objects. This fragmentation requires additional conversion steps and discards structure already present in the specification, such as subtask boundaries, ordering constraints, and quantitative measures. We introduce TBTCL (Temporal Behavior Tree Centric Learning), a framework...

    arxiv.org/abs/2611.13993 · PDF

  41. 41

    Universal Approximation Theorem for Metaplectic Neural Operators

    Ahmed Abdeljawad, Malek Yacine Rezoug

    cs.LG · math.FA

    We prove universal approximation for metaplectic neural operators on nonempty bounded Lipschitz domains. The nonlocal branches are whole-space metaplectic multipliers with zero extension and restriction, associated with any fixed free symplectic matrix. Assuming a smooth, globally Lipschitz, nonpolynomial activation, we obtain compact-uniform approximation of continuous operators from $C^s$ to $C^r$ and from $H^s$ to $H^r$ in the respective...

    arxiv.org/abs/2611.13991 · PDF

  42. 42

    RB-LDC: Redundancy-Balanced Latent Coding for Noise-Robust Diffusion Latents

    Hyunseok Jeong, Jaeho Jeon, Young-Sik Kim

    cs.LG · cs.CV

    Latent diffusion models decode approximately correct latents, making decoder sensitivity to perturbations relevant to image quality. After whitening, forward diffusion corrupts every direction at the same signal-to-noise ratio (SNR), despite unequal decoder sensitivities. We study an invertible linear layer between the encoder and diffusion process that shifts each direction's log-SNR. Its linear minimum mean-square-error (LMMSE) recovery...

    arxiv.org/abs/2611.13955 · PDF

  43. 43

    Geometry-Dependent Approximation and Stability Properties of Shallow ReLU$^k$ Networks on Bounded Euclidean Domains

    Yuqing Liu, Tong Mao, Jinchao Xu

    cs.LG · math.NA

    On bounded Euclidean domains, we study how hidden-parameter geometry affects shallow ReLU$^k$ networks through results on approximation rates, saturation-type lower bounds, Gram and stiffness matrix spectra, Bernstein and Marcinkiewicz--Zygmund inequalities, discrete least-squares recovery, and inverse approximation. Our analysis uses smooth localization and dyadic Fourier estimates. We quantify parameter geometry by the fill distance $h$ and...

    arxiv.org/abs/2611.13952 · PDF

  44. 44

    Chem World: A Large-Scale Benchmark and Physics-Aware Framework for Trustworthy Molecular Property Prediction

    Tianyou Bai, Huan Wang, Mingchen Gao, Tongrui Yang, Fangyue Lin, Pinze Ren, Zhenlin Zhao, Siming Dong

    cs.LG · cs.AI

    Molecular property prediction is fundamental to computational chemistry and drug formulation research, yet existing benchmarks remain fragmented across datasets, properties, and evaluation protocols, limiting systematic evaluation of model generalization. We introduce Chem World, a large-scale benchmark integrating 17 public datasets with over 800,000 molecular and chemical-system instances across 10 property tracks, covering single-molecule...

    arxiv.org/abs/2611.13945 · PDF

  45. 45

    PARDIG: Global Precipitation Estimates Learned from Sparse Satellite Observations

    Guy Shalev, Stephan Rasp, Boris Babenko, Andrew El-Kadi, Mohammed Alewi Hassen, Samier Merchant, Fred Zyda, Dominic...

    cs.LG

    Estimating precipitation at the global scale is vital for Earth system science and disaster response, yet observational coverage and measurement precision remain constrained. Existing solutions provide options that trade off desirable traits such as skill, latency, geographical coverage, temporal coverage, and spatial and temporal resolution. We present PARDIG, an artificial intelligence framework based on a spherical graph-mesh transformer...

    arxiv.org/abs/2611.13941 · PDF

  46. 46

    Dual-Core Tensor Wheel Decomposition with Structure-Cached Acceleration for Tensor Recovery

    Wenchao Xie, Qingsong Wang

    cs.LG

    Tensor wheel decomposition captures complex correlations in high-dimensional data through interactions between outer-ring factors and a high-order central core. However, updating these components requires the explicit construction and manipulation of large-scale matrices, which limits computational efficiency and incurs particularly high computational and memory costs for the central-core update. To address these limitations, we propose...

    arxiv.org/abs/2611.13938 · PDF

  47. 47

    Learning from First Discrepancies with Joint Reset and Pilot Guarantees

    Michal Čertík

    cs.LG

    A run that stops at its first wrong decision shows the queries and their correct labels up to and including that decision, and nothing after it. We study how many such runs are needed for learning. Each run tests a hypothesis, the pilot, chosen beforehand from a fixed binary class of VC dimension at most $d$, on a fresh record of at most $D$ queries drawn from one unknown distribution that does not depend on the pilot. The terminal...

    arxiv.org/abs/2611.13913 · PDF

  48. 48

    Target Selection with a Disappearing-Stimulus Mixed-Reality P300 Interface: Feasibility and Object-Distinguishable Versus Uniform Cues

    Runqi Pan, Takuto Niinuma, Yunshan Huang, Masato Sugino, Kenta Shimba, Kiyoshi Kotani

    cs.LG · cs.HC · eess.SP

    Mixed-reality (MR) P300 interfaces can register selectable destinations, such as wheelchair goals, directly in the user's surroundings. When the cues disappear between flashes, however, the user must retain an intended target that is no longer visible. We tested whether such an interface supports target selection and whether giving each cue a stable colour-shape identity improves it compared with visually uniform spheres. Six healthy adults...

    arxiv.org/abs/2611.13906 · PDF

  49. 49

    Conformal Risk Control for QoT-Aware Lightpath Feasibility Under Technology-Driven Domain Shift

    Kiarash Rezaei, Omran Ayoub, Paolo Monti, Carlos Natalino

    cs.LG

    Artificial intelligence and machine learning (AI/ML)-based regression models support operational decision-making in optical networks but typically expose only point predictions, which carry no statistically controlled information about their reliability. This paper proposes a post-hoc conformal risk control (CRC) framework that keeps a pretrained AI/ML regressor fixed, provides a finite-sample guarantee on uncertainty band miscoverage, and...

    arxiv.org/abs/2611.13893 · PDF

  50. 50

    Are Causal Neuron Circuits Unique and Transferable?

    Ömer Tekin Yavuz

    cs.LG · cs.CL

    Interpretability demonstrations often show that intervening on a group of neurons changes a model's answer. A recent example removes an 18-neuron group in Qwen3-32B and the model answers "Simba" instead of "Scar" to a two-hop Lion King riddle. Such a demonstration establishes that the group matters, but not how much of it is necessary, whether other groups could do the same, or whether the effect carries over to other inputs. We ask: when...

    arxiv.org/abs/2611.13885 · PDF

  51. 51

    Mechanistic Interpretability of Deep Neural Networks: Methods, Frameworks, and Perspectives

    Lorenzo Cirillo, Irene Amerini, Andrea Cavallaro

    cs.LG · cs.AI · cs.CV

    Deep neural networks underpin fundamental tasks in a wide range of signal domains, such as image, audio, and language processing. However, their internal mechanisms are notably difficult to interpret. Internal activations are structured signals transformed across layers that may encode a multitude of interdependent relationships, as individual neurons frequently represent multiple underlying concepts. Mechanistic interpretability aims to...

    arxiv.org/abs/2611.13868 · PDF

  52. 52

    Where to Approximate in Neurosymbolic Inference?

    Samy Badreddine, Emile van Krieken, Luciano Serafini, Antonio Vergari

    cs.LG

    Probabilistic neurosymbolic methods rely on weighted model counting (WMC) to combine neural predictors with symbolic constraints. Exact WMC is #P-hard and typically scales poorly, so many methods resort to approximations. By unifying existing approaches under a three-step bottom-up approximate compilation pipeline, we study where approximation budget is best allocated. Inspired by tensor networks, we instantiate two steps with tensor train...

    arxiv.org/abs/2611.13862 · PDF

  53. 53

    JAX-WAVE: A Differentiable Physics-Supervised Framework for Inverse Wave Design in Heterogeneous Media

    Honghui Du, Binyao Guo, Yushu Pan

    cs.LG

    Designing a prescribed wave field in a heterogeneous medium requires an initial pressure distribution that compensates for the medium's propagation response. Repeated numerical inversion solves this task afresh for each target and environment. We introduce JAX-WAVE for amortized inverse wave design: a shared neural network maps desired terminal fields and known wave-speed maps directly to bounded initial pressures. Training through a...

    arxiv.org/abs/2611.13826 · PDF

  54. 54

    AnchorMTS: Multi-View Signal Representation and Forecast-Anchored Fusion for Long-Term Time-Series Forecasting

    Gao Wu, Tingting Xu, Changning Wu, Tengfei Long, Yong Liu, Jiabin Jin, Kexin Zhang

    cs.LG

    Long-term time-series forecasters typically optimize a single numerical representation, even though the same history exposes complementary geometry, recurrence, and semantic structure. We propose AnchorMTS, which derives self-aligned visual and textual views from each observed window and uses them only to refine a complete numerical forecast. A trainable evidence path produces a candidate forecast, while a parameter-free output projection...

    arxiv.org/abs/2611.13794 · PDF

  55. 55

    Omni-JEV: Text and Dual-Stream Speech Decisions

    Shengzhou Kong

    cs.LG

    A user utterance does not uniquely determine the next action in full-duplex dialogue: its consequences depend on the assistant's response state and interaction history. Speech activity, utterance completion, and interruption intent provide related evidence, whose implications for a response require an explicit action policy. We formulate this task at collected-segment endpoints and introduce Omni-JEV, which maps text or role-indexed user and...

    arxiv.org/abs/2611.13777 · PDF

  56. 56

    Beyond Interaction Scores: Selecting Feature Interactions via Independent Replication

    Jincheng Gao, Shuo Sun

    cs.LG · cs.DB · stat.ML

    Learning feature interactions is essential for accurate prediction and interpretable modeling, yet a high interaction score may reflect fitting noise rather than a reproducible pattern. Ranking methods score feature pairs but leave the cutoff to the user; testing methods set the cutoff but rely on repeated model refits or a model of the joint feature distribution. We present LENS, a model-agnostic framework for selecting feature interactions...

    arxiv.org/abs/2611.13765 · PDF

  57. 57

    Mixed Signals: Weight-Space Programming for EEG and ECG

    Dara Varam, Taraneh Aminosharieh Najafi, David Atienza

    cs.LG · cs.AI · eess.SP

    Weight-space programming allows language models to read a trained neural network's weights and edit them. This opens the door to ``training'' by reasoning rather than through gradient descent. Prior work has demonstrated its functionality and promise for game-playing networks and fixing specific failures in language models, even extending to diffusion models. In this work, we present its applicability to models trained on biosignal data,...

    arxiv.org/abs/2611.13758 · PDF

  58. 58

    When Covariance Is Not Enough: Kernel-Based Regularization for Single-Cell Counterfactuals

    Korbinian Traeuble, Edward de Brouwer, Archit Verma, Doron Haviv, Kipper Fletez-Brant

    cs.LG · q-bio.GN

    Counterfactual editing of single cells, predicting how an observed cell would look had one factor been different, enables in-silico screening where experiments are infeasible, but is useful only if the edit is faithful to the individual and not merely to the condition. Concept bottleneck generative models make the effects of interventions interpretable by decomposing a cell's representation into known and unknown factors. Previous work uses a...

    arxiv.org/abs/2611.13748 · PDF

  59. 59

    Plug-and-Play ADMM for Inverse Problems with Flow Matching Denoiser

    Dibyanshu Kumar, Magda Gregorova

    cs.LG

    Plug-and-Play (PnP) methods with generative priors have achieved strong performance in inverse problems, with recent flow matching models enabling fast and scalable reconstruction. However, a key challenge in PnP methods is maintaining consistency between data fidelity updates and the learned generative prior, as data-consistency steps can move iterates away from the flow trajectory. We propose an ADMM-based PnP framework for flow matching...

    arxiv.org/abs/2611.13742 · PDF

  60. 60

    Deterministic, Reproducible Inference in Exact Integer Arithmetic, and Verifiable Distributed Inference as Its Consequence

    Sayed Amir Karim

    cs.LG · cs.DC · cs.SE

    Optimized floating-point inference is not reproducible across environments: the same model on different hardware, compilers or thread counts can produce different bits. We present tritify, an open-source runtime that executes the forward pass of real models -- decoder-only language models, Mixture-of-Experts decoders, and the Segment Anything 2.1 image/video stack -- with the entire run-time path in exact integer arithmetic; floating point is...

    arxiv.org/abs/2611.13730 · PDF

  61. 61

    Weighted-AoI-Driven UAV Trajectory and Reporting for Dynamic CKM Tracking

    Zeyu Zhao, Yueling Che, Sheng Luo

    cs.LG

    Existing channel knowledge map (CKM) construction assumes quasi-static environments, failing when channels rapidly decorrelate. This paper investigates continuous online dynamic CKM tracking over an airborne 2D plane via a UAV fleet. We formulate a long-term error minimization problem under UAV mobility and reporting constraints. This problem, however, is intractable with unobservable errors for unmeasured cells. To tackle this, we propose a...

    arxiv.org/abs/2611.13726 · PDF

  62. 62

    TRUER: Test-Time Reward Correction Using an Error-Aware Rectifier for Reinforcement Learning

    Mingxiang Shen, Jinhao Li, Jiaming Xu, Yibo Han, Guohao Dai, Yijia Zhang, Ningyi Xu

    cs.LG

    Test-time reinforcement learning enables language models to adapt using rewards derived from their own answers. However, majority voting can reinforce incorrect solutions even when correct alternatives are available. Our training audit shows that false positives (incorrect responses receiving positive rewards) dominate reward errors. We focus on a practical regime where ground-truth rewards are difficult to obtain at scale, while a limited...

    arxiv.org/abs/2611.13714 · PDF

  63. 63

    FedVP: Federated Learning across Heterogeneous Models by Pooling Locally Trained Verifiers

    Yifan Li, Xi Chen

    cs.LG

    Conventional federated learning relies on the shared model backbone to aggregate client updates, limiting collaboration among institutions that deploy large language models (LLMs) from different families and at different scales. We propose Federated Verifier Pool (FEDVP), a new federated learning paradigm that enables heterogeneous clients to collaborate through shared verification rather than parameter aggregation. Each client trains a...

    arxiv.org/abs/2611.13708 · PDF

  64. 64

    Bootstrapped On-Policy Distillation

    Yue Wu, Xueyu Zhou, Yangrong Hu, Jian Huang

    cs.LG · cs.AI

    On-policy distillation (OPD) improves language models by distilling teacher predictions on student-generated trajectories, while on-policy self-distillation (OPSD) removes the need for an external teacher by using a reference-conditioned copy of the model as the teacher. Despite its great early improvements, we find that OPSD does not remain consistently effective under continued training and can eventually suffer substantial performance...

    arxiv.org/abs/2611.13680 · PDF

  65. 65

    SepsisFlow: an event-stream foundation model for continuous risk warning and trajectory modelling in ICU sepsis

    Yawen Zhang, Wanzhe Xu, Zhenhua Zhang, Jinyu Wu, Yizhou Zhuang, Yitao Yang, Kenta Nakai

    cs.LG

    Sepsis is a leading cause of ICU death with heterogeneous onset and course, yet most warning models predict only whether sepsis will occur, leaving clinicians uncertain of its onset context and early trajectory once an alarm sounds. To answer these questions on one hourly clock from ICU admission, we built SepsisFlow: before onset, it combines a Mamba model pretrained on 2.57 million EHRs with about 1,700 hourly features in gradient-boosted...

    arxiv.org/abs/2611.13661 · PDF

  66. 66

    CP-MLPs: A Tensor-Rank Theory of Tied and Untied MLP Blocks

    Md Rifat Arefin, Farzaneh Heidari, Irina Rish, Guillaume Rabusseau

    cs.LG · stat.ML

    Transformer feed-forward blocks combine several design choices: squared activations, gating, and tied versus untied parameter sharing, whose separate contributions to expressivity remain poorly understood. We introduce CP-MLPs, a tensor framework in which each multiplicative unit contributes a rank-1 interaction with three roles: detecting an input feature, reading a value, and writing to the output. Tied gated blocks such as ReLU$^2$ force...

    arxiv.org/abs/2611.13647 · PDF

  67. 67

    MouldIt: Data Efficiency of Multi-Task Temporal Representation Learning in Injection Moulding

    Abdullah Ayad

    cs.LG · cs.RO · eess.SP

    Smart industrial solutions that can address multiple tasks require models that understand the underlying processes. We examine representation learning for modeling the injection moulding process as an example of such solutions. We consider its data efficiency and multi-modal benefit compared to classical machine learning alternatives. To this end, we synthesize a process-informed dataset based on a real reference cycle of an Erlanger Träger...

    arxiv.org/abs/2611.13646 · PDF

  68. 68

    Multifidelity Operator Inference for Non-intrusive Model Reduction under Data Scarcity

    Vivian Zhang, Tomoki Koike, Atticus Rex, Elizabeth Qian

    cs.LG · math.NA · stat.CO

    Many-query analyses, in which a computational model is repeatedly evaluated, play an important role in uncertainty quantification and optimization for engineering design. However, many-query analyses can be prohibitively expensive when considering high-fidelity numerical models, especially those governed by partial differential equations (PDEs). Surrogate modeling replaces costly simulations with cheap approximations, enabling practical...

    arxiv.org/abs/2611.13624 · 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.