cs.LG · 2026-08-10 · No. 80

Machine Learning, 2026-08-10.

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

    Diffusion LLMs as Targets and Adversaries: Mechanistic Safety Exploits

    Elena Dumitrescu, Gert Lek, Lydia Y. Chen, Jérémie Decouchant

    cs.LG · cs.AI

    Diffusion Large Language Models (DLLMs) replace autoregressive next-token prediction with iterative parallel denoising, yet their internal safety mechanisms remain poorly understood. In this work, we investigate DLLMs both as targets and as adversaries, exposing mechanistic vulnerabilities in diffusion-based alignment. We first show that safety alignment in DLLMs remains sparse and transferable across architectures. DLLMs initialized from...

    arxiv.org/abs/2608.07430 · PDF

  2. 02

    Beyond Myopic World Models: Long-Horizon End-to-End Training for Direct Future Prediction

    Xinyi Li, Zaishuo Xia, Chenjie Hao, Yubei Chen

    cs.LG

    World models are expected to support imagination over extended temporal horizons, yet most are still trained through local few-step prediction objectives and deployed by recursively rolling out their own predictions. This creates a fundamental mismatch: few-step losses optimize local transition fidelity, while long-horizon prediction depends on how errors and gradients propagate through the entire trajectory. As a result, transitions with...

    arxiv.org/abs/2608.07420 · PDF

  3. 03

    Beyond Post-Hoc Temperature Scaling: Bilevel Optimization for LLM Calibration

    Ruochen Jin, Zhanliang Wang, Zongyu Dai, Jiancong Xiao, Bojian Hou

    cs.LG

    Preference alignment often makes large language models (LLMs) overconfident and poorly calibrated. Traditional post-hoc temperature scaling is inherently domain-dependent: a temperature fitted on one domain does not generalize across domains. This motivates us to modify model parameters during training to improve calibration. We propose maximizing the entropy of predictive distributions as the calibration objective, which directly targets...

    arxiv.org/abs/2608.07419 · PDF

  4. 04

    FedDOSE: Federated Learning Framework Decomposing Site Effects for Modeling Brain Dynamic Functional Connectivity

    Deepank Girish, Yi Hao Chan, Yubin Zheng, Sukrit Gupta, Jagath C. Rajapakse

    cs.LG · eess.SP · q-bio.NC

    Functional Magnetic Resonance Imaging ( fMRI ) data are often pooled into collaborative multi-site consortia, as deep learning models for analyses require large datasets to generalize well. While Federated Learning (FL) offers a privacy-preserving paradigm for collaborative training, standard approaches continue to struggle with statistical heterogeneity. In particular, site differences pose a key challenge in multi-site data settings....

    arxiv.org/abs/2608.07393 · PDF

  5. 05

    Omni-modal decomposition autoencoders learn full-stack wearable disentangled representations

    Ioannis Ziogas, Ensieh Khazaei, Bilal Taha, Aamna Al Shehhi, Ahsan H. Khandoker, Leontios J. Hadjileontiadis,...

    cs.LG · cs.AI · eess.AS · eess.SP · stat.ML

    Learning disentangled representations is a key requirement for developing versatile, general-purpose, and sustainable models in multi-modal wearable computing. However, existing approaches do not operate as full-stack wearable processors, i.e., they do not simultaneously address task-specific classification performance, disentangled and interpretable representation learning, fusion, and generative modeling of highly heterogeneous multi-modal...

    arxiv.org/abs/2608.07385 · PDF

  6. 06

    Trajectory-Relative Hindsight Distillation for Agentic Reinforcement Learning

    Haoyu Zheng, Yun Zhu, Qing Wang, Wenqiao Zhang

    cs.LG · cs.CL

    Recent agentic reinforcement learning methods use hindsight to complement sparse outcome rewards. However, a completed rollout can yield many such signals, leaving their appropriate allocation across turns unclear. We introduce TRIAL, a trajectory-relative hindsight distillation framework with a unified turn-aligned scoring protocol. For each decision turn, TRIAL extracts an outcome view of that decision's realized consequence and evaluates...

    arxiv.org/abs/2608.07371 · PDF

  7. 07

    Residual Algebra for Representation-Preserving Learning

    Yao Wu

    cs.LG

    Learning from heterogeneous representations is usually reduced to feature concatenation, which erases which representation produced an error. We instead algebraize the residual: a representation is a typed object that owns both a coordinate system and the residual it leaves unresolved, and learning is an ordered composition of operators that preserve or deliberately erase that type. Fold realizes the objects as point-in-time conditional-mean...

    arxiv.org/abs/2608.07349 · PDF

  8. 08

    Aftab: A Comprehensive Benchmark of CNN Encoders and Advanced Value Functions in Parallelized Q-Networks

    Taha Shieenavaz, Shabnam Zareshahraki, Loris Nanni

    cs.LG · cs.AI

    Recent advancements in deep reinforcement learning have increasingly favored simplified, highly parallelized paradigms. Notably, the Parallelized Q-Network (PQN) algorithm achieves stable off-policy learning without relying on computationally expensive replay buffers or target networks. However, the representational capacity and parameter efficiency of visual encoders operating in these buffer-free settings remain underexplored. In this work,...

    arxiv.org/abs/2608.07335 · PDF

  9. 09

    When GNNs Fail: Quantifying and Overcoming Temporal Correlation Volatility in Time Series

    Chen Shao, Yue Wang, Zhenyi Zhu, Zhanbo Huang, Tobias Käfer, Zonghan Wu, Danai Koutra

    cs.LG

    Modeling multivariate time series by representing them as graphs, where individual series act as nodes and pairwise temporal corre- lations serve as edges, has gained significant traction. Recent advances in Graph Neural Networks (GNNs) have demonstrated strong perfor- mance by assuming a static graph topology and aggregating information from neighboring series. In this work, we investigate the representa- tional power of GNNs for forecasting...

    arxiv.org/abs/2608.07333 · PDF

  10. 10

    Is SwiGLU's Open Positive Tail Necessary? Evidence from Closed-Tail Gating with MemGLU

    Yuting Ge, Pengju Yang, Mingkai Nie

    cs.LG

    We test whether decoder-only language-model FFNs require SwiGLU's open positive tail. We introduce MemGLU as a closed-tail comparator derived from a memristive branch geometry. Across paired 9M and 30M pretraining runs with three seeds, MemGLU remains within about 0.1% of SwiGLU in validation NLL. Trained SwiGLU checkpoints are sensitive to positive-tail suppression, while mechanism diagnostics show that the two models use their gates...

    arxiv.org/abs/2608.07323 · PDF

  11. 11

    From Optimal Actions to World Models: Identifiability of Transition Kernels in Discounted MDPs

    Neal Batra

    cs.LG

    We study what can be recovered about the transition probabilities of a Markov decision process from optimal actions alone. This is closely related to the inverse problem considered by Letcher et al., who ask when the dynamics can be recovered from numerical \(Q\)-values. Here the numerical values themselves are not observed; only the optimal actions are known, for every reward in a given class. For state-action rewards \(r(s,a)\), knowing the...

    arxiv.org/abs/2608.07301 · PDF

  12. 12

    FUSE: Feature-Wise Unified Specialization with Cross-Column Exchange for Mixed-Type Tabular Flow Matching

    Suman Cha, Seongchan Lee, Dohyun Ko, Hyunjoong Kim

    cs.LG · cs.AI

    Generating mixed-type tabular data requires jointly modeling diverse feature distributions and their complex cross-column dependencies. Variational flow matching handles distinct endpoints via factorized distributions, yet leaves feature-specific processing and cross-column interactions implicit within a shared backbone. We introduce Feature-wise Unified Specialization with cross-column Exchange (FUSE) to explicitly separate these roles. FUSE...

    arxiv.org/abs/2608.07294 · PDF

  13. 13

    A foundation-model approach to pediatric headache classification from rs-fMRI

    Guilherme S. Imai Aldeia, Clara Moon, Julie Shulman, Navil Sethna, Allison Smith, Alyssa Lebel, William G. La Cava,...

    cs.LG

    Headache is the most common neurological disorder in children and substantially affects quality of life. We investigated whether resting-state functional MRI (rs-fMRI) can support pediatric headache classification using machine learning. We encoded rs-fMRI data using NeuroSTORM, a recent foundation model, and fine-tuned it to distinguish healthy controls from children with headache and subsequently classify headache subtypes. We compared...

    arxiv.org/abs/2608.07287 · PDF

  14. 14

    TOFD: Target-Oriented Feature Decoupling against Poisoning Attacks in Split Federated Learning

    Yuhan Xie, Jingrong Huang, Chen Lyu

    cs.LG · cs.AI

    Split Federated Learning (SFL) facilitates privacy-preserving collaborative training with reduced client-side overhead. However, its split architecture introduces unique attack surfaces, rendering it vulnerable to diverse poisoning attacks. Most existing defenses fail to exploit the split paradigm, limiting their ability to detect and contain malicious behaviors at an early stage. To bridge this gap, we propose Target-Oriented Feature...

    arxiv.org/abs/2608.07274 · PDF

  15. 15

    Learning Suffers More Than the Policy Class Under Partial Observability: A Closed-Form Analysis

    Idil Gözel

    cs.LG · math.OC

    When a reinforcement learning agent cannot observe the full state, we usually blame its policies: it cannot see enough to represent a good one. We show that in a solvable case the bigger problem lies elsewhere. Even when a good policy is available and the agent's value function is expressive enough to describe it exactly, learning still ends up somewhere far worse. We study a partially observed linear-quadratic problem in which a standard...

    arxiv.org/abs/2608.07228 · PDF

  16. 16

    Stochastic Autoregressive Learning

    Ilan Doron-Arad, Idan Mehalel, Elchanan Mossel

    cs.LG

    Motivated by LLMs, which generate outputs by iteratively sampling from next-token distributions, we introduce a PAC-learning model for binary stochastic autoregressive learning. This generalizes the deterministic autoregressive learning framework of Joshi et al., COLT 2025. In our model, one fixed generator assigns a Bernoulli next-token distribution to every prompt string. Starting from an input prompt, a token is sampled and appended to the...

    arxiv.org/abs/2608.07224 · PDF

  17. 17

    An AI4AI Framework for Visual Token Pruning

    Zhen Liu, Wenli Huang, Wei Song, Yuhan Liu, Zhiqin Yang, Jingwen Fu

    cs.LG · cs.CV

    Visual-token pruning can substantially reduce the inference cost of multimodal large language models (MLLMs), yet existing methods largely rely on fixed, handcrafted heuristics and costly expert trial and error. As pruning objectives, budgets, and model architectures diversify, manually navigating the expanding design space becomes increasingly difficult. This paper aims to build an AI4AI framework for visual-token pruning by addressing a...

    arxiv.org/abs/2608.07193 · PDF

  18. 18

    MAUPITI: On-Device Prototype-Based Learning on a Smart Infrared Sensor

    Beatrice Alessandra Motetti, Tanguy Dugas du Villard, Matteo Risso, Alessio Burrello, Francesco Daghero, Enrico...

    cs.LG

    Low-resolution infrared (IR) array sensors represent an interesting solution for privacy-preserving human sensing in embedded systems. In this letter, we describe a smart multi-pixel IR sensor integrating a 16$\times$16 thermal MOSFET (TMOS) array and a RISC-V microcontroller extended with low-precision SIMD instructions, capable of on-device learning and continual adaptation for pose and gesture recognition tasks under tight memory and power...

    arxiv.org/abs/2608.07192 · PDF

  19. 19

    Conformal Fusion Under Missing Modalities

    Alireza Moayedikia

    cs.LG

    Multimodal fusion architectures typically assume all modalities are available at inference, yet sensor failures, acquisition variability, and cost constraints routinely produce incomplete observations. Existing work treats modality absence as a prediction-accuracy problem, leaving a more basic question unanswered: whether a model's confidence estimates remain calibrated when an entire input stream is removed. We argue that missing-modality...

    arxiv.org/abs/2608.07183 · PDF

  20. 20

    Momba: Network Modernization Improves Multi-Objective Reinforcement Learning

    Adam Štafa, Santeri Heiskanen, Petr Novotný, Joni Pajarinen

    cs.LG · cs.AI

    Recent advances in deep reinforcement learning (RL) have shown that improving neural network architectures can yield substantial gains in sample efficiency and asymptotic performance without altering the underlying algorithms. In contrast, work on multi-objective reinforcement learning (MORL), which aims to discover a set of policies that balance trade-offs among conflicting objectives, has predominantly focused on algorithmic innovations,...

    arxiv.org/abs/2608.07180 · PDF

  21. 21

    Fluid-DiT: Graph-Free Diffusion Transformers for Fluid Flow Simulations Learning

    Shentong Mo, Guolin Ke

    cs.LG · cs.AI · cs.CE

    Simulating complex fluid flows requires capturing full equilibrium distributions rather than just mean trajectories, yet high-fidelity solvers remain computationally prohibitive. Recent advances, such as Diffusion Graph Networks (DGNs), have combined diffusion models with graph neural networks to sample equilibrium states directly from unstructured meshes, enabling distributional accuracy even from short simulations. However, graph-based...

    arxiv.org/abs/2608.07161 · PDF

  22. 22

    Edge Sparsification via Temporal Forman-Ricci Curvature for Dynamic Graph Learning

    Poupak Azad, Cuneyt Gurcan Akcora, Kiarash Shamsi

    cs.LG

    Temporal graph learning has become essential for analyzing real-world systems whose interactions continuously evolve over time, including financial transaction networks, communication systems, and online social platforms. However, learning from large-scale temporal graphs remains computationally challenging when networks are dense and rapidly changing. To address this limitation, we propose a network-curvature-inspired edge sparsification...

    arxiv.org/abs/2608.07158 · PDF

  23. 23

    Capacity Confounds and Coverage Guarantees in Adaptive Sub-model Federated Learning

    Alireza Moayedikia, Alicia Troncoso Lora

    cs.LG · cs.DC

    Sub-model federated learning lets resource-constrained clients train width-reduced versions of a global model, but existing methods allocate capacity by device resources alone. A natural next step, allocating capacity by each client's data heterogeneity as estimated from the updates the server already observes, has been repeatedly suggested. We ask whether that step is possible, using HAS-FL, an adaptive capacity-allocation framework, as a...

    arxiv.org/abs/2608.07157 · PDF

  24. 24

    Machine Learning-Based Inter-Crystal Scatter Recovery for Ultra-High Resolution PET Imaging

    Alexandre Bernier, Roger Lecomte, Jean-Baptiste Michaud

    cs.LG · physics.med-ph

    Inter-crystal scatter (ICS) events pose a significant challenge in ultrahigh- resolution positron emission tomography (UHR-PET), especially as detector crystals become smaller and their readouts increasingly segmented. Current approaches either reject these events, reducing sensitivity, or accept them with suboptimal positioning algorithms, degrading image resolution. We present a feed forward neural network to optimize ICS event recovery by...

    arxiv.org/abs/2608.07155 · PDF

  25. 25

    Interpretable reinforcement learning with decision-tree pruning

    Mark Leon Ringer, Michel Tokic

    cs.LG · cs.AI

    Reinforcement learning policies are difficult to inspect, but interpreting them is a prerequisite for trustworthiness. Converting a trained policy into explicit decision-tree rules improves transparency and the resulting artifacts often remain too complex for human understanding. We present a pruning process that simplifies such rule-based policies while preserving task performance and making edits to the policy auditable. The process defines...

    arxiv.org/abs/2608.07151 · PDF

  26. 26

    Online Conformal Prediction Beyond Feedback

    Joar Skalse, Edoardo Pona, Osvaldo Simeone, Nicola Paoletti

    cs.LG

    Uncertainty quantification is essential when deploying machine learning models in safety-critical applications. Online conformal prediction (OCP) provides theoretically principled uncertainty quantification for arbitrary black-box classifiers and non-i.i.d. data streams by constructing prediction sets that are guaranteed to contain the true label at a user-specified frequency. OCP usually updates prediction sets using feedback from previously...

    arxiv.org/abs/2608.07139 · PDF

  27. 27

    Modular TTT: Rethinking Test-Time Training as Composable Modules

    Bohao Tang, Zhen Qin, Yuqi Pan, Zheng Li, Pengfei Liu, Ya Zhang

    cs.LG · cs.CL

    Test-time training (TTT) views sequence modeling as an online learning problem in which fast weights are updated by an internal learning rule. Despite the growing number of TTT variants, existing approaches typically hard-code each variant separately, which makes it difficult to design new TTT methods and to isolate the role of each component. To address this, we propose Modular TTT, a framework that represents the inner learner as a directed...

    arxiv.org/abs/2608.07110 · PDF

  28. 28

    Synthetic LiDAR Data Generation and Deterministic Downsampling for Point Cloud Classification on the Edge

    Niclas Meyer, Stefan Reitmann

    cs.LG · cs.RO

    Deploying three-dimensional deep learning frameworks to low-power embedded processors is bottlenecked by the unstructured nature of spatial data and the resource-intensive distance sorting algorithms often used before neural network inference. To address this gap, this paper presents a hardware-constrained workflow optimized for native execution on the Raspberry Pi 5. To account for the reality gap between noiseless, clean computer-aided...

    arxiv.org/abs/2608.07106 · PDF

  29. 29

    Beyond Isolation: Unlocking Reinforcement Learning Component Synergy for Sample-Efficient Continuous Control

    Qi Zhao, Guozheng Ma, Yilun Kong, Lu Li, Haoyu Wang, Zilin Wang, Tiantian Zhang, Yuxing Wang, Jian Sha, Yongzhe...

    cs.LG · cs.AI

    Reinforcement learning systems are significantly more complex than other machine learning paradigms due to inherent properties, causing RL system design to jointly account for many tightly coupled factors. Despite advances in individual algorithmic components, their functional interdependencies remain underexplored: do they exhibit mutual synergy or counterproductive interference? To bridge this gap, we conduct a systematic investigation and...

    arxiv.org/abs/2608.07086 · PDF

  30. 30

    Accounting Graph Transformer for Short-History Multi-KPI Forecasting in Small Businesses

    Shrutendra Harsola, Vignesh Subrahmaniam

    cs.LG · cs.AI

    Small businesses often have only 12-24 months of accounting history, yet planning and risk workflows require coordinated forecasts across financial statements. We study joint 12-month forecasting of 13 income-statement, balance-sheet, cash-flow, and working-capital key performance indicators (KPIs) from 71 monthly ledger series. We introduce the Accounting Graph Transformer (AGT), which represents each ledger series as a masked token,...

    arxiv.org/abs/2608.07037 · PDF

  31. 31

    Hyperbolic Graph Embedders for Link Prediction and Topology Reconstruction

    Robert Jankowski, Maksim Kitsak, Dorota Celińska-Kopczyńska

    cs.LG · cs.SI · physics.soc-ph

    Hyperbolic embeddings provide compact geometric representations of complex networks in hyperbolic spaces, but systematic comparisons of methods developed in machine learning, network science, and algorithmics remain rare. We benchmark 13 unsupervised hyperbolic graph embedders under a unified protocol for link prediction and topology reconstruction on synthetic and empirical networks. The protocol captures both missing-link recovery and the...

    arxiv.org/abs/2608.07029 · PDF

  32. 32

    Every Cache Entry Earns Its Place: Global Allocation of Resolution and Coverage for KV Cache Compression

    Haolin Tian, Yuzhe Liu, Tonghan Wang

    cs.LG

    As large language models (LLMs) process increasingly long contexts, KV cache storage and repeated access have become a major bottleneck. Existing KV cache compression methods rely on predefined, fixed compression rules and are typically developed around either token eviction or merging. As a result, cache resources can neither flow freely across layers, heads, and context slots, nor be jointly allocated to balance local resolution and...

    arxiv.org/abs/2608.07001 · PDF

  33. 33

    Beyond Foundation Models: Dimension-Aware Neural Architecture Search with Small-Data Representation Models for Cryocooler Lifetime Prediction

    Gregor Molan, Grafika Jati, Francesco Barchi, Andrea Acquaviva, Aljaž Osterman, Martin Molan

    cs.LG · cs.AI

    Large-scale pretrained time-series models achieve strong results through large-scale pretraining and task-agnostic representation learning, but they rely on abundant, diverse data that industrial and scientific domains often lack. We therefore propose the FSD-RM (Family of Small-Data Representation Models) paradigm as a practical alternative for limited, domain-specific telemetry. Rather than relying on large-scale pretraining, we focus on...

    arxiv.org/abs/2608.06993 · PDF

  34. 34

    Density-aware Hierarchical Clustering Based on Element-Categorized Connection Subgraphs

    Yuning Yu, José Rodríguez-Piñeiro, Xuefeng Yin, Bin Feng

    cs.LG · cs.AI

    Clustering is a fundamental data mining technique for pattern recognition through unsupervised learning. Among various clustering methods, hierarchical clustering, density-based clustering, and graph clustering stand out as representative approaches. For hierarchical clustering, it can be categorized into agglomerative and divisive modes to construct clusters in a recursive manner. The key aspect of both modes is the calculation of...

    arxiv.org/abs/2608.06990 · PDF

  35. 35

    How Molecular Generative Models Organize Molecular Identity

    Raul Ortega-Ochoa, Tejs Vegge, Jens S. Bakander, Luis Mantilla Calderon, Alan Aspuru-Guzik, Tonio Buonassisi

    cs.LG · physics.chem-ph

    Generative models for matter are often evaluated as samplers over output representations, and their latent spaces are commonly used as proxies for navigating chemical space. Much less is known about how these models internally arrange discrete chemical identities within those representations. We study this arrangement by making molecular identity explicit and pulling it back through the generative process. Through these pullbacks we probe the...

    arxiv.org/abs/2608.06956 · PDF

  36. 36

    A Rate Separation for Agnostic Direct Sums

    Mihir More, Aritra Das, Debayan Gupta

    cs.LG

    Hanneke, Moran, and Waknine \cite{HannekeMoranWaknine2024} asked how the agnostic PAC learning curve of the direct sum $C^r$ depends on the single-instance learning curve $\epsagn(n\mid C)$ and on $r$. We show that the single-instance learning rate does not determine the direct-sum rate. Let $\F$ be the class of the two constant binary functions and let $\G$ consist of the zero function and the identity function. Both classes have agnostic...

    arxiv.org/abs/2608.06951 · PDF

  37. 37

    ELMZip: Onboard Satellite Image Compression via Extreme Learning Machines for Efficient Downlink

    Woojin Cho, Junghwan Park, Sangcheol Sim, Steve Andreas Immanuel, Junhyuk Heo, Darongsae Kwon

    cs.LG · cs.CV

    The acquisition of multispectral imagery via small satellites (e.g., CubeSats) presents significant data downlink challenges due to high data volumes and restricted communication windows. While onboard image compression is critical to address this bottleneck, traditional methods often struggle to adapt to the nonlinear statistics of multi-band, multi-resolution data. To overcome these limitations, we propose ELMZip, a novel framework based on...

    arxiv.org/abs/2608.06942 · PDF

  38. 38

    Walkable to Whom? Capturing Subjective Variability in Walkability Perception Using Multimodal Deep Learning

    Moloud Damandeh, Meead Saberi

    cs.LG · cs.CV

    Visual perception of walkability varies substantially across individuals, reflecting differences in personal characteristics, experiences, and preferences. Existing studies, however, often reduce these diverse judgements to aggregated scores, implicitly assuming uniform perception, and commonly rely on vehicle-mounted street-view imagery that does not reflect the pedestrian's visual experience. This paper introduces a dataset of 29,870...

    arxiv.org/abs/2608.06934 · PDF

  39. 39

    MiCoPro: End-to-End Mixed Precision HW/SW Co-design with HW-aware Proxy Model

    Zijun Jiang, Yangdi Lyu

    cs.LG

    Quantized Neural Networks~(QNN) with low-bitwidth data have proven promising in efficient storage and computation on edge devices. To mitigate accuracy degradation while maximizing speedup, layer-wise mixed-precision quantization~(MPQ) becomes a popular solution. However, existing algorithms for exploring MPQ schemes are limited in flexibility and efficiency. Comprehending the complex impacts of different MPQ schemes on post-training...

    arxiv.org/abs/2608.06916 · PDF

  40. 40

    Recent advances in weakly supervised learning: New supervision paradigms, assumption relaxations, and practical solutions

    Wei Wang, Gang Niu, Masashi Sugiyama

    cs.LG

    Deep learning has achieved great success in recent years thanks to the availability of high-quality, well-annotated training data. However, this requirement is often not met in real-world applications. Weakly supervised learning aims to train an accurate model with incomplete, inexact, or inaccurate supervision. In this chapter, we will discuss recent advances in this field, including new supervision paradigms, relaxed assumptions, and...

    arxiv.org/abs/2608.06896 · PDF

  41. 41

    PRISM: Principled Reference Identification for Schrodinger Bridge Model

    Forouzan Fallah, Yezhou Yang

    cs.LG

    Schrödinger bridge models restore a clean signal from a degraded observation by following the conditional bridges of a reference process, yet this reference is chosen heuristically, typically white noise with a hand-tuned schedule. We develop PRISM, a theory of bridge reference design. We characterize the time-varying Gaussian references that remain exactly tractable with per-mode schedules: precisely those whose instantaneous covariances...

    arxiv.org/abs/2608.06893 · PDF

  42. 42

    SkillAligner: Treating Retrieved Skills as Adaptable Drafts at Execution Time

    Qinfeng Li, Dalin He, Yuntai Bao, Ying Yang, Ruoxi Chen, Xinyan Yu, Lizhou Liang, Ge Su, Wenqi Zhang, Xuhong Zhang

    cs.LG

    General-purpose skills promise reusable procedural knowledge for language agents, yet semantic relevance does not guarantee execution utility: a retrieved skill may encode assumptions that conflict with the current task, execution environment, or other retrieved skills. We formalize this problem as the skill--execution misfit. To address it, we propose SkillAligner, a training-free execution-time skill adaptation framework that treats...

    arxiv.org/abs/2608.06880 · PDF

  43. 43

    Bridging the Gap Between Hyperdimensional Computing and Kernel Methods via the Nyström Method

    Quanling Zhao, Anthony Hitchcock Thomas, Ari Brin, Xiaofan Yu, Tajana Rosing

    cs.LG · cs.AI

    Hyperdimensional computing (HDC) is an approach from the cognitive science literature for solving information processing tasks using data represented as high-dimensional random vectors. The technique has a rigorous mathematical backing, and is easy to implement in energy-efficient and highly parallel hardware like FPGAs and "processing-in-memory" architectures. The effectiveness of HDC in machine learning largely depends on how raw data is...

    arxiv.org/abs/2608.06860 · PDF

  44. 44

    Mathematical Principles and Experimental Discoveries of the Emergence of Symbolic Patterns in Artificial Neural Networks

    Quanshi Zhang, Qihan Ren, Siyu Lou

    cs.LG

    Artificial Neural networks (ANNs) are often treated as black-box models, making explainability a central challenge in deep learning. Many engineering methods have been proposed to approximately explain the ANN from various perspectives, such as feature attribution and visualization. However, it remains a long-standing open question whether the complex inference logic of an ANN can be explained exhaustively and concisely as sparse symbolic...

    arxiv.org/abs/2608.06839 · PDF

  45. 45

    Graph Machine: Exploring Edge Mechanisms as an Inductive Bias

    Lintai Hou

    cs.LG

    Transformers provide a powerful architecture for global content-based matching, but reasoning problems may benefit from a stronger inductive bias toward iterative traversal of latent relations. We introduce Graph Machine, an architecture with two explicit edge-based mechanisms: Edge-augmented attention, in which edges modulate attention between nodes, and edge-centric referral, in which nodes exchange addresses to update their edges....

    arxiv.org/abs/2608.06834 · PDF

  46. 46

    Multiscale Reward Hedging from Correct Demonstrations

    Pahan Dewasurendra

    cs.LG

    Learning from correct demonstrations is harder than supervised learning when many answers are correct: after predicting, the learner sees one valid answer but not whether its own answer was valid, nor any reward. Existing reward-hedging guarantees consequently assume a finite reward class. We give the first horizon-free guarantee for continuous classes. The key is to hedge in one shared vote over tolerant optimality tests at every accuracy...

    arxiv.org/abs/2608.06825 · PDF

  47. 47

    Understanding Differentiable Embeddings Through Differential and Integral Geometry

    Xinyu Zhang, Klaus Mueller

    cs.LG

    How can an analyst decide whether a nonlinear dimensionality reduction embedding can be trusted? Existing diagnostics provide only partial answers: projection glyphs characterize local sensitivity, map-continuity scores measure local conditioning, and transport-based analyses reveal path-dependent inconsistencies. However, these methods appear unrelated and provide no common framework for understanding when they agree or not. We show that...

    arxiv.org/abs/2608.06809 · PDF

  48. 48

    Faster Query-Key Learning Sharpens Attention in Self-Attention Models

    Rahul Vashisht, Harish G. Ramaswamy

    cs.LG

    A standard self-attention layer consists of two interacting circuits: the query-key circuit that governs attention allocation, and the output-value circuit that maps attended representations to predictions. Collapsed and factorized parameterizations of the query-key and output-value circuits lead to qualitatively different attention patterns. In particular, some parameterizations give sharper attention to task-relevant tokens, at a similar...

    arxiv.org/abs/2608.06776 · PDF

  49. 49

    ArchEGraph: A Large-Scale Graph Dataset for Geometry-Topology-Physics Aligned Building Energy Modeling

    Yihui Li, Yihui Chen, Kaidi Zha, Xiaoyue Yan, Zhexuan Yu, Shiqi Dai, Jun Xiao, Jun Yin, Ramon Elias Weber, Borong Lin

    cs.LG

    Accurate estimation of building energy use is essential for achieving carbon neutral and sustainable buildings. To better understand the influence of design decisions on building energy use and calibrate machine learning models that can give architects and engineers rapid design feedback, large-scale datasets are needed that explicitly map building geometry to performance. We present ArchEGraph, a large-scale benchmark dataset that represents...

    arxiv.org/abs/2608.06772 · PDF

  50. 50

    Hidden Gauge Controls Feature Specialization in ReLU Networks

    Tongxi Wang

    cs.LG · cs.AI

    Training changes a network's predictions while allocating task-relevant structure across its internal units. In an overparameterized ReLU network, several neurons can begin with exactly the same functional role, yet one may acquire a teacher feature while the others become redundant. We call the identity of that neuron feature ownership and ask whether it can be controlled by a parameter choice invisible to the initial predictor. In a...

    arxiv.org/abs/2608.06766 · PDF

  51. 51

    CubicQuant: Parametric Non-Uniform Codebooks for High-Throughput LLM Inference with 1-8-Bit Weights

    Xuetian Gao

    cs.LG · cs.DC

    Weight quantization for large-language-model inference must balance adaptive reconstruction levels with representations regular enough for efficient GPU execution. Uniform integers constrain each group to a linear grid. Low-bit floating-point formats use a fixed exponent-mantissa structure, while learned codebooks gain flexibility at the cost of irregular decoding and additional metadata. We introduce CubicQuant, a parametric non-uniform...

    arxiv.org/abs/2608.06763 · PDF

  52. 52

    Sub-Quadratic Bisimulation Metrics via Approximate Nearest Neighbors: Coverage-Augmented Guarantees and Computable Two-Sided Certificates

    Ibne Farabi Shihab, Joyanta Jyoti Mondal

    cs.LG

    Bisimulation metrics quantify behavioral similarity in Markov decision processes, but their Wasserstein fixed-point operator updates every state pair and incurs quadratic pairwise work. We give a certificate-carrying sub-quadratic method for MDPs with bounded transition support and a useful low-dimensional indexing representation: an approximate-nearest-neighbor index selects the pairs updated by the exact restricted operator, while monotone...

    arxiv.org/abs/2608.06762 · 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.