Online Learning-to-Defer with Varying Experts
arXiv:2605.12340v1 Announce Type: cross Abstract: Learning-to-Defer (L2D) methods route each query either to a predictive model or to external experts. While existing work studies this problem in batc
Knowledge catalogue
arXiv:2605.12340v1 Announce Type: cross Abstract: Learning-to-Defer (L2D) methods route each query either to a predictive model or to external experts. While existing work studies this problem in batc
arXiv:2605.11199v1 Announce Type: cross Abstract: Trained lattice samplers are usually judged by the ensembles they generate. Here we instead analyze the trained field-space function itself: a flow-ma
arXiv:2605.12235v1 Announce Type: cross Abstract: We study optimal policy learning under combined budget and minimum coverage constraints. We show that the problem admits a knapsack-type structure and
arXiv:2605.11291v1 Announce Type: new Abstract: In this paper, we provide a computable characterization of the geometry of optimal representations in Contrastive Learning (CL) when the classes are imb
arXiv:2605.11172v1 Announce Type: new Abstract: We introduce SODA, a generalization of Optimistic Dual Averaging, which provides a common perspective on state-of-the-art optimizers like Muon, Lion, Ad
arXiv:2605.11977v1 Announce Type: new Abstract: We present a unified framework for 3D geometric abstraction using a single continuous 4D wire, parameterized as a B-spline with spatial coordinates and
arXiv:2605.12419v1 Announce Type: new Abstract: Despite the rapid advancements in large language model (LLM) development, fine-tuning them for specific tasks often results in the catastrophic forgetti
arXiv:2605.12446v1 Announce Type: cross Abstract: Large language models (LLMs) often produce answers with high certainty even when they are incorrect, making reliable confidence estimation essential f
arXiv:2602.12139v2 Announce Type: replace Abstract: Transformers excel at time series modelling through attention mechanisms that capture long-term temporal patterns. However, they assume uniform time
arXiv:2605.11803v1 Announce Type: new Abstract: As Video Large Language Models (Video-LLMs) scale to longer and more complex videos, their inference cost grows rapidly due to the large volume of visua
arXiv:2605.11570v1 Announce Type: new Abstract: Activation functions are what make deep networks expressive: without them, the model collapses to a linear map. Yet we still evaluate training mostly fr
arXiv:2605.12345v1 Announce Type: new Abstract: Parameter-efficient fine-tuning (PEFT) techniques offer task-specific fine-tuning at a fraction of the cost of full fine-tuning, but require separate fi
arXiv:2605.11459v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models achieve remarkable flexibility and generalization beyond classical control paradigms. However, most prevailing VLA
arXiv:2605.11525v1 Announce Type: new Abstract: Missing values are routinely treated as defects to be eliminated through deletion or imputation prior to machine learning. In many applied domains, howe
arXiv:2506.23619v2 Announce Type: replace-cross Abstract: This paper investigates the impact of posterior drift on out-of-sample forecasting accuracy in overparametrized machine learning models. We do
arXiv:2605.11178v1 Announce Type: new Abstract: Neural Sheaf Diffusion (NSD) generalizes diffusion-based Graph Neural Networks by replacing scalar graph Laplacians with sheaf Laplacians whose learned
arXiv:2605.12199v1 Announce Type: new Abstract: Emergent misalignment (EM), where fine-tuning on a narrow task (like insecure code) causes broad misalignment across unrelated domains, was first demons
arXiv:2605.12313v1 Announce Type: new Abstract: Multi-hop question answering (QA) remains a significant challenge in the biomedical domain, requiring systems to integrate information across multiple s
arXiv:2605.12072v1 Announce Type: new Abstract: Dropout-based sparse-view 3D Gaussian Splatting (3DGS) methods alleviate overfitting by randomly suppressing Gaussian primitives during training. Existi
arXiv:2602.01418v2 Announce Type: replace Abstract: We propose Parabolic Position Encoding (PaPE), a parabola-based position encoding for vision modalities in attention-based architectures. Given a se
arXiv:2605.10953v1 Announce Type: cross Abstract: The demand for high-resolution subsurface imaging and continuous Earth monitoring has driven rapid growth in active and passive seismic data from dens
arXiv:2605.11684v1 Announce Type: new Abstract: We propose a Byzantine-resilient federated conformal prediction (FCP) method that leverages partial model sharing, where only a subset of model paramete
arXiv:2602.04042v2 Announce Type: replace Abstract: We propose Partition Tree, a novel tree-based framework for conditional density estimation over general outcome spaces that supports both continuous
arXiv:2510.05497v5 Announce Type: replace-cross Abstract: Large-scale Mixture of Experts (MoE) Large Language Models (LLMs) have recently become the frontier open-weight models, achieving remarkable m
arXiv:2605.11427v1 Announce Type: new Abstract: 4D Gaussian Splatting (4DGS) enables high-quality dynamic novel view synthesis, yet current models remain monolithic bitstreams that clients must downlo
arXiv:2510.02107v4 Announce Type: replace Abstract: AdaBoost sequentially fits so-called weak learners to minimize an exponential loss, which penalizes misclassified data points more severely than oth
arXiv:2605.11730v1 Announce Type: new Abstract: Automated red-teaming for LLMs often discovers narrow attack slices, missing diverse real-world threats, and yielding insufficient data for safety fine-
arXiv:2605.11266v1 Announce Type: new Abstract: Recent advances in Gaussian Splatting have enabled fast, high-fidelity 3D scene generation, yet these methods remain purely visual and lack an understan
arXiv:2602.13690v2 Announce Type: replace Abstract: Magnetic-anomaly navigation, leveraging small-scale variations in the Earth's magnetic field, is a promising alternative when GPS is unavailable or
arXiv:2512.05683v2 Announce Type: replace Abstract: Optical aberrations significantly degrade image quality in microscopy, particularly when imaging deeper into samples. These aberrations arise from d
arXiv:2605.11346v1 Announce Type: new Abstract: Physics-informed deep learning (PIDL) neural networks have shown their capability as a useful instrument for transportation practitioners in utilizing t
arXiv:2602.08058v2 Announce Type: replace Abstract: In the presence of occlusions and measurement noise, geometrically accurate scene reconstructions -- which fit the sensor data -- can still be physi
arXiv:2605.12492v1 Announce Type: new Abstract: We introduce Pion, a spectrum-preserving optimizer for large language model (LLM) training based on orthogonal equivalence transformation. Unlike additi
arXiv:2605.11225v1 Announce Type: cross Abstract: Large language model (LLM)-based agents frequently generate seemingly coherent plans that fail upon execution due to infeasible actions, constraint vi
arXiv:2212.02011v3 Announce Type: replace Abstract: Point cloud learning is receiving increasing attention. However, most existing point cloud models lack the practical ability to deal with the unavoi
arXiv:2605.11594v1 Announce Type: new Abstract: High-fidelity reconstruction of driving scenes is crucial for autonomous driving. While recent feedforward 3D Gaussian Splatting (3DGS) methods enable f
arXiv:2605.11520v1 Announce Type: new Abstract: Unsupervised point cloud segmentation is critical for embodied artificial intelligence and autonomous driving, as it mitigates the prohibitive cost of d
arXiv:2602.15473v2 Announce Type: replace Abstract: Gradient-based optimizers are highly sensitive to design choices in their adaptive learning rate mechanisms. To address this limitation, we introduc
arXiv:2605.11497v1 Announce Type: new Abstract: Zero-shot skeleton-based action recognition (ZSSAR) is typically treated as a skeleton-text alignment problem: encode joint-coordinate sequences, align
arXiv:2605.12144v1 Announce Type: new Abstract: In visual localization, Absolute Pose Regression (APR) enables real-time 6-DoF camera pose inference from single images, yet critically depends on fine-
arXiv:2512.11883v3 Announce Type: replace-cross Abstract: Over-aligning image generation models to a generalized aesthetic preference conflicts with user intent, particularly when 'anti-aesthetic' out
arXiv:2605.11511v1 Announce Type: cross Abstract: The validity of statistical inference depends critically on how data are collected. When data gathered through active data collection (ADC) are reused
arXiv:2605.11652v1 Announce Type: cross Abstract: We study posterior contraction rates for sparse Bayesian Kolmogorov-Arnold networks (KANs) over anisotropic Besov spaces, providing a statistical foun
arXiv:2605.12411v1 Announce Type: cross Abstract: AI agents negotiate and transact in natural language with unfamiliar counterparts: a buyer bot facing an unknown seller, or a procurement assistant ne
arXiv:2605.12422v1 Announce Type: new Abstract: Automatic generation of educational materials using large language models (LLMs) is becoming increasingly common, but assigning difficulty levels to suc
arXiv:2605.11303v1 Announce Type: new Abstract: We investigate the use of Large Language Models (LLMs) for zero-shot prediction of Ryff Psychological Well-Being (PWB) scores from spontaneous speech. U
arXiv:2605.11453v1 Announce Type: cross Abstract: Practitioners deploying multi-agent large language model (LLM) systems must currently choose between communication topologies such as chain, star, mes
arXiv:2605.12160v1 Announce Type: new Abstract: Vision-Language-Action (VLA) policies are typically evaluated as if the user had finished typing or speaking before the robot begins acting. In real dep
arXiv:2605.12243v1 Announce Type: new Abstract: Conversational scams, such as romance and investment scams, are emerging as a major form of online fraud. Unlike one-shot scam lures such as fake lotter
arXiv:2605.11363v1 Announce Type: cross Abstract: Presentation generation is moving beyond static slide creation toward end-to-end presentation video generation with research grounding, multimodal med
arXiv:2605.12382v1 Announce Type: new Abstract: Large language models (LLMs) exhibit systematic preferences for well-known entities, a phenomenon often attributed to popularity bias. However, the exte
arXiv:2605.12241v1 Announce Type: cross Abstract: Specialized foundation models are beginning to emerge in various medical subdomains, but pretraining methodologies and parametric scaling with the siz
arXiv:2605.11535v1 Announce Type: new Abstract: Existing work on linear constrained Markov decision processes (CMDPs) has primarily focused on stochastic settings, where the losses and costs are eithe
arXiv:2605.11299v1 Announce Type: cross Abstract: Code generation is typically trained in the primal space of programs: a model produces a candidate solution and receives sparse execution feedback, of
arXiv:2605.10984v1 Announce Type: new Abstract: Uncertainty quantification complements model predictions by characterizing their reliability, which is essential for high-stakes decision making such as
arXiv:2605.11506v1 Announce Type: new Abstract: Score-based diffusion models achieve state-of-the-art performance for inverse problems, but their practical deployment is hindered by long inference tim
arXiv:2601.13780v3 Announce Type: replace Abstract: Graph diffusion models achieve state-of-the-art performance in graph generation but suffer from quadratic complexity in the number of nodes -- and m
arXiv:2605.12289v1 Announce Type: new Abstract: Leveraging the rich world knowledge of Large Language Models (LLMs) to enhance Reinforcement Learning (RL) agents offers a promising path toward general
arXiv:2605.11608v1 Announce Type: new Abstract: Comparing post-training LLM variants, such as quantized, LoRA-adapted, and distilled models, requires a diagnostic that identifies how a variant has dri
arXiv:2605.12260v1 Announce Type: new Abstract: Long-horizon language agents accumulate conversation history far faster than any fixed context window can hold, making memory management critical to bot