Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2602. 03570v2 Announce Type: replace Abstract: Audio-visual joint representation learning under Cross-Modal Generalization (CMG) aims to transfer knowledge from a labeled source modality to an unlabeled target modality through a unified discrete representation space.
arXiv:2607. 22944v1 Announce Type: cross Abstract: Invariants, the relations expected to hold among measured signals of a network, underpin applications from verification to traffic generation, telemetry imputation, and input validation, yet writing them by hand demands rare expertise in both formal logic and networking.
arXiv:2607. 24453v1 Announce Type: cross Abstract: Learning from minimal human supervision is a long-standing goal in medical image analysis, where dense expert annotations are costly.
arXiv:2502. 16184v3 Announce Type: replace Abstract: The EU Artificial Intelligence Act (AIA) establishes different legal principles for different types of AI systems.
arXiv:2607. 22665v1 Announce Type: new Abstract: Machine unlearning is becoming increasingly critical in the context of data privacy regulations, particularly for recommendation systems that are directly trained on user interaction data.
arXiv:2607. 23649v1 Announce Type: new Abstract: Differential privacy provides formal privacy guarantees for training neural networks on sensitive data, while Bayesian deep learning offers a principled framework for uncertainty-aware prediction.
arXiv:2607. 24645v1 Announce Type: cross Abstract: The wide-scale use of sparse autoencoders (SAEs) as interpretability tools is limited by inconsistent links between SAE features and model behavior.
arXiv:2607. 23310v1 Announce Type: cross Abstract: We study an online variant of discrete fair division under generalized assignment budget constraints.
arXiv:2510. 16077v2 Announce Type: replace-cross Abstract: Domain Incremental Learning (DIL) is a sub-branch of continual learning that aims to address the never-ending arrival of new domains without catastrophic forgetting.
arXiv:2607. 24057v1 Announce Type: new Abstract: Real-world Reinforcement Learning depends on the ability to formulate safety constraints into a policy.
arXiv:2507. 21134v2 Announce Type: replace-cross Abstract: As large language models (LLMs) are increasingly deployed in high-risk domains such as law, finance, and medicine, systematically evaluating their domain-specific safety and compliance becomes critical.
arXiv:2607. 23940v1 Announce Type: new Abstract: Physics-Informed Neural Networks (PINNs) solve differential equations by minimizing the residual of a nonlinear operator over a neural parameterization of the solution.
arXiv:2605. 07407v2 Announce Type: replace Abstract: We show that information can be transferred post-hoc across independently trained health foundation models (FMs), each pretrained on ~20M minutes of wearable sensor data from ~172K participants, by aligning their data-dependent coordinate systems.
arXiv:2607. 23869v1 Announce Type: cross Abstract: Randomized features provide a scalable approximation to kernel machines, but their performance depends strongly on the choice of feature distribution.
arXiv:2607. 23777v1 Announce Type: cross Abstract: The discovery of scaling laws has motivated training neural networks on ever increasing quantities of data.
arXiv:2607. 24553v1 Announce Type: cross Abstract: Standardized echocardiography conclusions provide meaningful supervision for learning ECG representations of echocardiography-derived cardiac findings.
A growing number of applications, such as biometrics and retrieval-augmented generation (RAG), rely on cosine similarity scores computed between vector embeddings of text, images, or audio. These systems return similarity scores through their APIs for ranking and verification.
The wide-scale use of sparse autoencoders (SAEs) as interpretability tools is limited by inconsistent links between SAE features and model behavior. Features with clear activation descriptions may have weak or unexpected causal effects; steering can vary across prompts or oppose the intended direction; and activation-based feature selection can miss features that produce the desired output change.
Split learning enables collaborative model training by partitioning neural networks across clients and servers. However, improper split placement can lead to severe privacy leakage through intermediate representations.
Standardized echocardiography conclusions provide meaningful supervision for learning ECG representations of echocardiography-derived cardiac findings. Global ECG--text alignment may entangle modality-specific factors, while long-tailed finding distributions provide sparse positive supervision for low-prevalence conditions.