Knowledge catalogue
arXiv:2605.10061v1 Announce Type: cross Abstract: Futrell and Mahowald (2025) frame the success of neural language models (LMs) as supporting gradient, usage-based linguistic theories. I argue that LM
arXiv:2605.09160v1 Announce Type: new Abstract: Learned representations are often invariant to rotational transformations, leaving individual dimensions non-identifiable and interchangeable. We study
arXiv:2512.24552v2 Announce Type: replace Abstract: This paper proposes a novel second-order optimization algorithm based on the Optimal Control Principle (OCP), applicable to large-scale optimization
arXiv:2605.09200v1 Announce Type: new Abstract: We study adversarial noisy bandits given a known function class F. In each round, the adversary selects a function f in F, the learner chooses an arm, a
arXiv:2605.10722v1 Announce Type: new Abstract: Molecular Graph Neural Networks (GNNs) are increasingly common in drug discovery, particularly for Quantitative Structure-Activity Relationship (QSAR) s
arXiv:2605.10818v1 Announce Type: new Abstract: Periodic signals are critical for representing physical and perceptual phenomena. Scalar, real angular measures, e.g., radians and degrees, result in di
arXiv:2509.15816v5 Announce Type: replace Abstract: The Muon optimizer has demonstrated remarkable empirical success in handling matrix-structured parameters for training neural networks. However, a s
arXiv:2512.18248v3 Announce Type: replace Abstract: The low-rank adaptation (LoRA) algorithm for fine-tuning large models has grown popular in recent years due to its remarkable performance and low co
arXiv:2605.10775v1 Announce Type: cross Abstract: A surprising phenomenon in the training of neural networks is the ability of gradient descent to find global minimizers of the training loss despite i
arXiv:2601.22158v3 Announce Type: replace Abstract: Modern diffusion/flow-based models for image generation typically exhibit two core characteristics: (i) using multi-step sampling, and (ii) operatin
arXiv:2505.23617v3 Announce Type: replace-cross Abstract: Effective video tokenization is critical for scaling transformer models for long videos. Current approaches tokenize videos using space-time p
arXiv:2605.09565v1 Announce Type: new Abstract: We consider the problem of learning an unknown subset N_ext{target} of a domain in an online setting. In each round t, the learner predicts a set of ite
arXiv:2605.10040v1 Announce Type: new Abstract: The rapid evolution of generative paradigms has enabled the creation of highly realistic imagery, which escalating the risks of identity fraud and the d
arXiv:2511.01292v2 Announce Type: replace-cross Abstract: Pretrained Transformers can perform in-context learning (ICL) from a few demonstrations, but this ability can fail sharply when the test distr
arXiv:2605.09454v1 Announce Type: cross Abstract: We study the extit{single-index bandit} problem, where rewards depend on an unknown one-dimensional projection of high-dimensional contexts through an
arXiv:2605.09991v1 Announce Type: new Abstract: Mode connectivity has been widely studied, yet the role of the optimizer remains underexplored. We revisit it through optimizer-induced implicit regular
arXiv:2605.09813v1 Announce Type: cross Abstract: We investigate the control and optimization of vertical federated learning (VFL), a class of distributed machine learning (ML) methods in which edge/f
arXiv:2605.09656v1 Announce Type: new Abstract: Recent advances in artificial intelligence (AI) have enabled effective perception and language models for robots, but their deployment remains computati
arXiv:2603.22421v2 Announce Type: replace Abstract: Predicting long-term bone remodeling after mandibular reconstruction would be of great clinical benefit, yet standard generative models struggle to
arXiv:2605.10161v1 Announce Type: new Abstract: Weight decay remains one of the most widely used regularization mechanisms for training convolutional neural networks, yet it is still commonly applied
arXiv:2605.08955v1 Announce Type: new Abstract: We develop and evaluate a data-driven approach for detecting unusual (anomalous) patient-management decisions using past patient cases stored in electro
arXiv:2506.01301v2 Announce Type: replace Abstract: Theory-of-Mind (ToM) enables humans to infer mental states-such as beliefs, desires, and intentions-forming the foundation of social cognition. Howe
arXiv:2605.09976v1 Announce Type: new Abstract: Online Temporal Action Localization (On-TAL) aims to detect the occurrence time and category of actions in untrimmed streaming videos immediately upon t
arXiv:2605.08328v1 Announce Type: cross Abstract: Generative models based on flow matching have emerged as a powerful paradigm for inverse problems, offering straighter trajectories and faster samplin
arXiv:2605.08371v1 Announce Type: new Abstract: Visual Geometry Transformer (VGGT) is a strong feed-forward model for multiple 3D tasks, but its Alternating-Attention (AA) stack scales quadratically i
arXiv:2605.09036v1 Announce Type: new Abstract: Accurate and efficient storm-surge emulation is essential for coastal hazard assessment, yet high-fidelity hydrodynamic models remain too expensive for
arXiv:2605.08632v1 Announce Type: cross Abstract: Speculative decoding accelerates Large Language Models (LLMs) inference by using a lightweight draft model to propose candidate tokens that are verifi
arXiv:2605.10791v1 Announce Type: new Abstract: Knowledge Graph Question Answering (KGQA) aims to answer user questions by reasoning over Knowledge Graphs (KGs). Recent KGQA methods mainly follow the
arXiv:2605.09339v1 Announce Type: cross Abstract: Human color categories are not uniformly distributed in perceptual space, yet most computational color models still assume fixed and evenly structured
arXiv:2605.08121v1 Announce Type: cross Abstract: Early detection of plant diseases is critical for improving crop productivity, while it also facilitates the foundations of precision agriculture. Rec
arXiv:2605.09319v1 Announce Type: new Abstract: With the proliferation of AI-generated images, digital watermarking has become an essential safeguard for protecting intellectual property and mitigatin
arXiv:2605.09043v1 Announce Type: cross Abstract: One partner says 'Fine' meaning resolution; the other hears surrender. The word is shared; the affective uptake is not. We formalize this as affective
arXiv:2605.09552v1 Announce Type: cross Abstract: Recently, Muon and related spectral optimizers have demonstrated strong empirical performance as scalable stochastic methods, often outperforming Adam
arXiv:2605.08915v1 Announce Type: new Abstract: Deep learning paradigms, such as PINNs and neural operators, have significantly advanced the solving of PDEs. However, they often struggle to capture th
arXiv:2605.08945v1 Announce Type: new Abstract: Action quality assessment (AQA) aims to automatically quantify the execution quality of human actions in videos and is valuable for applications such as
arXiv:2605.10922v1 Announce Type: new Abstract: Recent advances in 3D generative models have rapidly improved image-to-3D synthesis quality, enabling higher-resolution geometry and more realistic appe
arXiv:2605.08388v1 Announce Type: new Abstract: Human-AI teams play a pivotal role in improving overall system performance when neither the human nor the model can achieve such performance on their ow
arXiv:2605.10310v1 Announce Type: new Abstract: Existing alignment research is dominated by concerns about safety and preventing harm: safeguards, controllability, and compliance. This paradigm of ali
arXiv:2605.08574v1 Announce Type: new Abstract: As synthetic images become increasingly realistic, reliable synthetic image detection techniques are of pressing need to prevent their misuse. Despite s
arXiv:2605.10335v1 Announce Type: cross Abstract: Adaptive optimizers, most notably Adam, have become the default standard for training large-scale neural networks such as Transformers. These methods
arXiv:2605.09189v1 Announce Type: new Abstract: The scaling laws guiding modern model training were calibrated for a single regime: data-rich, single-epoch pretraining. The dominant such scaling law f
arXiv:2605.08308v1 Announce Type: cross Abstract: Wi-Fi sensing detects human motions and activities by analysing the channel state information (CSI) derived from Wi-Fi transmissions. However, the imp
arXiv:2605.09154v1 Announce Type: new Abstract: We introduce a predictive model that estimates the pre-training loss of large models from model size (N), batch size (B) and number of weight updates (K
arXiv:2605.09044v1 Announce Type: new Abstract: Deep continual learning requires models to adapt to new tasks without retraining from scratch. However, neural networks can lose their ability to adapt
arXiv:2605.10278v1 Announce Type: new Abstract: Background: Radiogenomics allows identification of radiological biomarkers for genomic phenotypes. In glioblastoma, these biomarkers could potentially c
arXiv:2605.10713v1 Announce Type: cross Abstract: We study sparse recovery when observations come from mixed-quality sources: a small collection of high-quality measurements with small noise variance
arXiv:2605.08343v1 Announce Type: new Abstract: Institutions may benefit from collaborative inference on time-series data. In settings where privacy is necessary, multi-party computation (MPC) is a st
arXiv:2605.09071v1 Announce Type: new Abstract: Score Distillation Sampling (SDS) and its variants have been widely used for text-to-3D generation by distilling 2D image diffusion priors. However, the
arXiv:2605.10815v1 Announce Type: new Abstract: Audio-visual large language models (AVLLMs) have recently emerged as a powerful architecture capable of jointly reasoning over audio, visual, and textua
arXiv:2605.09850v1 Announce Type: cross Abstract: Post-hoc calibration is usually evaluated as a function of logits or softmax confidence alone, even as routing-augmented architectures increasingly ac
arXiv:2605.08858v1 Announce Type: new Abstract: Ante-hoc interpretability methods based on prototypes provide highly accurate explanations by utilizing the intuitive 'this looks like that' reasoning p
arXiv:2605.10409v1 Announce Type: new Abstract: Existing image simplification techniques often rely on Non-Photorealistic Rendering (NPR), transforming photographs into stylized sketches, cartoons, or
arXiv:2210.05108v2 Announce Type: replace-cross Abstract: We study projection-free methods for functional constrained optimization with convex or smooth nonconvex objectives. Such problems arise in ap
arXiv:2605.08585v1 Announce Type: cross Abstract: Deep learning models in medical imaging typically operate as parametric memory, diagnosing patients by recalling fixed knowledge learned during traini
arXiv:2509.21677v2 Announce Type: replace Abstract: We present Prophecy, a tool for automatically inferring formal properties of feed-forward neural networks. Prophecy is based on the observation that
arXiv:2510.10730v2 Announce Type: replace-cross Abstract: We provide a unified algorithmic framework for ensemble sampling in nonlinear contextual bandits and develop corresponding regret bounds for t
arXiv:2605.10748v1 Announce Type: cross Abstract: One-Shot Federated Learning, where a central server learns a global model in a single communication round, has emerged as a promising paradigm. Howeve
arXiv:2605.10458v1 Announce Type: new Abstract: Atomic properties such as partial charges or multipoles encode chemically meaningful information that can inform downstream molecular property predictio
arXiv:2504.14697v3 Announce Type: replace Abstract: The evolution of tokens through deep transformer models can be modeled as an interacting particle system that has been shown to exhibit an asymptoti
arXiv:2506.01404v2 Announce Type: replace Abstract: This paper introduces an innovative error feedback framework designed to mitigate quantization noise in distributed graph filtering, where communica