Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2607. 09973v1 Announce Type: cross Abstract: Industrial sound design requires audio generation systems that not only produce realistic audio, but also preserve the perceptual identity of a reference, support controllable variation, and remain efficient for practical workflows.
arXiv:2607. 11475v1 Announce Type: new Abstract: Safety alignment in large language models can be fragile under fine-tuning, as even benign task adaptation may increase harmful compliance.
arXiv:2509. 20114v3 Announce Type: replace Abstract: We study \emph{online episodic Constrained Markov Decision Processes} (CMDPs) under both stochastic and adversarial constraints.
arXiv:2607. 10112v1 Announce Type: cross Abstract: Safety alignment in large language models remains brittle across languages: prompts reliably refused in English can elicit harmful compliance in non-English and low-resource settings.
arXiv:2607. 10539v1 Announce Type: new Abstract: Existing approaches to infer user traits and generate responses consistent with a persona rely on static prompting.
arXiv:2607. 10233v1 Announce Type: cross Abstract: Melody skeleton extraction aims to derive a shorter melody that preserves structural notes while removing ornaments.
arXiv:2606. 24903v2 Announce Type: replace Abstract: Few-shot label acquisition lacks a label-free signal for when additional labels cease to improve accuracy: existing stopping criteria either require a held-out validation set (violating the few-shot premise) or rely on theoretically ungrounded heuristics, so we introduce the spectral saturation index $S(K)=\mathrm{erank}(\hat{\Sigma}_W^{(K)})/K$, the exponential spectral entropy of the pooled within-class covariance normalized by per-class support size $K$, which measures the exploration rate per label and falls below a fixed threshold $\tau=0.
arXiv:2607. 10555v1 Announce Type: cross Abstract: Generative Large Language Models (LLMs) have revolutionized information retrieval, yet their strictly parametric nature frequently leads to severe factual hallucinations when confronted with complex queries beyond their epistemic boundaries.
arXiv:2607. 11523v1 Announce Type: cross Abstract: When should an intelligent assistant speak up without being asked?
arXiv:2607. 10476v1 Announce Type: cross Abstract: Large language models (LLMs) have emerged as a powerful tool for retrieving knowledge through seamless, human-like interactions.
arXiv:2607. 11751v1 Announce Type: cross Abstract: As multi-agent, tool-using LLM systems are deployed, a common safety net is a runtime monitor that checks each message, tool call, or step on its own.
arXiv:2510. 09497v2 Announce Type: replace-cross Abstract: In endovascular surgery, endovascular interventionists push a thin tube called a catheter, guided by a thin wire to a treatment site inside the patient's blood vessels to treat various conditions such as blood clots, aneurysms, and malformations.
arXiv:2503. 08884v3 Announce Type: replace-cross Abstract: Unimodal vision models are known to rely on spurious correlations, but it remains unclear to what extent Multimodal Large Language Models (MLLMs) exhibit similar biases despite language supervision.
arXiv:2606. 12936v2 Announce Type: replace-cross Abstract: Wet-lab robots can improve the reproducibility, throughput, and safety of biomedical experiments, but scaling their learning requires customizable simulators for safe and reproducible task generation, open editable laboratory assets, and efficient pipelines that turn limited demonstrations into usable training data.
arXiv:2512. 07901v3 Announce Type: replace-cross Abstract: Von Neumann founded both game theory and the theory of self-reproducing automata, but the two programs never merged.
arXiv:2607. 11257v1 Announce Type: cross Abstract: Pathology Foundation Models (PFMs) offer powerful Whole Slide Image (WSI) representations but suffer from massive computational costs.
arXiv:2607. 10618v1 Announce Type: cross Abstract: We consider the recovery of a pair of sparse vectors from a limited number of nonlinear observations of their superposition: $y_i=g(\inner{\ba_i}{\bPhi\bw^\ast+\bPsi\bz^\ast})+e_i$, $i=1,\dots,m$, with $m\ll n$, incoherent orthonormal bases $\bPhi,\bPsi$, a scalar link $g$, and noise $e_i$ that may be heavy-tailed or contaminated.
arXiv:2607. 11128v1 Announce Type: cross Abstract: Real-time driving risk assessment provides an essential basis for proactive safety by identifying and quantifying the danger of ongoing road interactions before adverse outcomes occur.
arXiv:2607. 10951v1 Announce Type: new Abstract: We introduce Sticky Jump Diffusions (SJDs), continuous-time Markov processes on $\mathbb R^d$ whose discrete anchors are token embeddings.
arXiv:2607. 09787v1 Announce Type: cross Abstract: LiDAR semantic segmentation is a key perception task in autonomous driving, where false predictions can affect downstream planning and safety-critical decision-making.