Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2610.19252v1 Announce Type: cross Abstract: Fairness audits with several sensitive attributes are often performed one coordinate at a time. This can miss purely intersectional disparities: a sc...
arXiv:2610.19539v1 Announce Type: new Abstract: Aggregate performance metrics do not identify local modelling inaccuracies in machine learning (ML) surrogates used for thermal-system optimisation. Th...
arXiv:2610.18354v1 Announce Type: new Abstract: Game-theoretic benchmarks claim that large language model (LLM) strategic decision-making varies across game types, interaction structures, and capabil...
arXiv:2610.18638v1 Announce Type: new Abstract: Combining pretrained models is a common strategy for building AI systems, whether by aligning a vision encoder with a language model or merging fine-tu...
arXiv:2610.17085v1 Announce Type: cross Abstract: On many estimates, artificial general intelligence (AGI) is likely to emerge within the next few decades or even sooner. Once it does, it may quickly...
arXiv:2610.17281v1 Announce Type: cross Abstract: SE(3)-equivariance provides a useful geometric inductive bias for visuomotor policies, encoding that actions covary with rigid transformations of the...
arXiv:2610.19910v1 Announce Type: new Abstract: Large language model agents increasingly improve by distilling experience into reusable skills, but errors in that experience can persist in learned ru...
arXiv:2610.14077v1 Announce Type: cross Abstract: Embodied Audio-Visual Navigation (AVN) requires an agent to locate sound targets by jointly leveraging visual and auditory cues. While prior work has...
arXiv:2610.14293v1 Announce Type: cross Abstract: Video diffusion models increasingly serve as priors for 4D generation, but the generated videos often lack geometric consistency. Recent approaches a...
arXiv:2610.14443v1 Announce Type: cross Abstract: Model cards, system cards, and safety reports increasingly document LLM safety behaviors, yet safety evaluations typically start from predefined benc...
arXiv:2610.14474v1 Announce Type: cross Abstract: Medication recommendations seek to deliver personalized drug combinations based on patients' clinical information, where ensuring transparent decisio...
arXiv:2605.16343v2 Announce Type: replace-cross Abstract: Looped language models (LoopLMs) increase effective computational depth by repeatedly applying shared Transformer blocks without increasing t...
arXiv:2605.27016v2 Announce Type: replace-cross Abstract: Large language models (LLMs) can hallucinate, but do model uncertainty estimators reliably reveal when they do? Although uncertainty estimato...
arXiv:2610.20560v1 Announce Type: cross Abstract: 3D object composition is pivotal to digital content creation and robot simulation, while also serving as a critical testbed for physical intelligence...
arXiv:2610.17636v1 Announce Type: cross Abstract: The Safety-Aware Denoiser (SAD) is a recent, training-free, safety-guidance framework for text diffusion models. During inference, SAD steers text sa...
arXiv:2610.17856v2 Announce Type: cross Abstract: Time-series analysis is increasingly studied in multimodal settings, yet most benchmarks pair temporal signals with text or with visualizations deriv...
The paper introduces Counterfactual Semantic Saliency (CSS), a black-box, model-agnostic method that measures how the removal of objects from a scene shifts the semantic description produced by vision‑language models (VLMs). Using CSS, the authors compare prominent VLMs to a large human psychophysics baseline (16,603 responses across 307 scenes) and find that VLMs over‑rely on large, centrally located, and highly salient objects, while under‑rely on people compared to humans. The study highlights a significant scene‑comprehension gap between VLMs and human perception, with size bias emerging as a key factor driving model‑human divergence.
arXiv:2605.26937v3 Announce Type: replace-cross Abstract: Parametric knowledge in large language models (LLMs) is a cornerstone of their success, yet remains poorly understood. Existing knowledge ben...
arXiv:2610.16175v1 Announce Type: new Abstract: Real-world studies (RWS) analyze routinely collected healthcare data, such as electronic health records and insurance claims, to estimate treatment eff...
arXiv:2610.19731v1 Announce Type: new Abstract: Goal misgeneralization is a common failure mode in which an agent behaves competently out of distribution but does not pursue the intended goal. Howeve...