Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2606. 27536v1 Announce Type: cross Abstract: Speech emotion recognition (SER) often relies on hard consensus labels that collapse annotator disagreement.
arXiv:2601. 16406v2 Announce Type: replace-cross Abstract: Rare-event prediction is critical in domains such as healthcare, finance, reliability engineering, customer support, aviation safety, where positive outcomes are infrequent yet potentially catastrophic.
arXiv:2602. 01588v3 Announce Type: replace-cross Abstract: Multimodal time series forecasting is crucial in real-world applications, where decisions depend on both numerical data and contextual signals.
arXiv:2606. 27984v1 Announce Type: new Abstract: Multimodal feature fusion can effectively capture complex patterns in real-world data by integrating complementary information from different modalities.
arXiv:2506. 16150v4 Announce Type: replace-cross Abstract: As large language models (LLMs) advance, concerns about their misconduct in complex social contexts intensify.
arXiv:2606. 27752v1 Announce Type: new Abstract: Single-cell perturbation models can reduce costly wet-lab screening by predicting how cells respond transcriptionally to interventions.
arXiv:2601. 14302v2 Announce Type: replace-cross Abstract: Image transmission and processing systems in resource-critical applications face significant challenges from adversarial perturbations that compromise mission-specific object classification.
arXiv:2606. 28123v1 Announce Type: new Abstract: Last-iterate convergence and generalization guarantees in first-order convex learning hinge on the monotonicity of the update operator.
arXiv:2606. 26519v2 Announce Type: replace Abstract: Large language models (LLMs) can make scientific software easier to use.
arXiv:2606. 27922v1 Announce Type: cross Abstract: Current multimodal reflection mechanisms for long video understanding predominantly rely on closed-loop self-reflection within internal parameters.
arXiv:2606. 27973v1 Announce Type: cross Abstract: Speech-based cognitive impairment detection offers a noninvasive, accessible alternative to costly biomarker assays, yet transformer-based models remain clinically uninterpretable.
arXiv:2606. 28011v1 Announce Type: cross Abstract: We propose an agentic Large Language Model (LLM) framework for active Fault-Tolerant Control (FTC) that transforms fault detection outputs into constraint-aware recovery actions grounded in plant-specific knowledge.
arXiv:2511. 20196v2 Announce Type: replace Abstract: Multimodal large language models (MLLMs) can inadvertently memorize privacy-sensitive information during training.
arXiv:2606. 28050v1 Announce Type: cross Abstract: LLM-as-a-Judge and self-evaluation pipelines implicitly assume that evaluation is easier than generation.
arXiv:2605. 20282v3 Announce Type: replace-cross Abstract: Machine unlearning in Vertical Federated Learning (VFL) has attracted growing interest, yet existing methods certify forgetting solely using output-level metrics.
arXiv:2606. 25034v2 Announce Type: replace-cross Abstract: General-purpose models often struggle to reliably identify and understand real-world multimodal risks, largely due to the inherent multimodal adversarial nature of content and AI safety.
arXiv:2606. 27731v1 Announce Type: cross Abstract: Despite their strong general capabilities, large language models (LLMs) often remain unreliable when outputs must be numerically precise.
arXiv:2606. 27832v1 Announce Type: new Abstract: Statistical adversarial detection (SAD) treats detection as a two-sample test.
arXiv:2606. 28152v1 Announce Type: new Abstract: We propose KL-Coupled Policy Regularization (KCPR), a policy coordination framework for Reward-Punishment Reinforcement Learning (RPRL).
arXiv:2510. 24187v3 Announce Type: replace-cross Abstract: We consider the adversarial linear bandits setting and present a unified algorithmic framework that bridges Follow-the-Regularized-Leader (FTRL) and Follow-the-Perturbed-Leader (FTPL) methods, extending the known connection between them from the full-information setting.