Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2608. 20240v1 Announce Type: new Abstract: X-band SAR satellites (8-12 GHz) play a critical role in disaster response, environmental monitoring, and military intelligence.
arXiv:2607. 04234v2 Announce Type: replace-cross Abstract: Deformable object manipulation poses challenges beyond task completion: successful execution must also maintain safe physical interaction, holding the object stably without slip or drop while avoiding excessive deformation.
arXiv:2608. 19885v1 Announce Type: cross Abstract: Streaming systems that maintain a pool of expert models must repeatedly decide whether to reuse an existing expert for arriving data, spawn a new one, or defer.
arXiv:2608. 20117v1 Announce Type: new Abstract: The emergence of large-scale Weather and Climate (W&C) datasets offers new opportunities for modeling extreme Earth events (ExEE) and their impacts using deep learning.
arXiv:2608. 19206v1 Announce Type: cross Abstract: Contemporary Large Language Models (LLMs) are increasingly aligned to suppress hallucinations, prioritizing factual retrieval over combinatorial creativity.
arXiv:2604. 02567v2 Announce Type: replace-cross Abstract: Despite the growing use of generative artificial intelligence (GenAI) in entrepreneurship, research on its impact remains fragmented.
arXiv:2604. 06501v2 Announce Type: replace Abstract: Analogical reasoning is a hallmark of human intelligence, enabling us to solve new problems by transferring knowledge from one situation to another.
arXiv:2608. 19973v1 Announce Type: cross Abstract: Recently, open-vocabulary 3D object detection (3D-OVD) has gained increasing attention for its ability to detect unseen objects in 3D scenes.
arXiv:2608. 19536v1 Announce Type: cross Abstract: Learning-based global point cloud registration has achieved remarkable progress, yet its reliance on geometric representations makes existing methods sensitive to variations in point density, scan pattern, viewpoint, and sensor characteristics.
The paper introduces DICS, a clustering-based framework that uses data-informed priors to construct a compact set of candidate splits for decision tree classifiers. By incorporating class-aware structure, DICS reduces the split search space, preserving predictive performance while cutting training time. The authors provide theoretical analysis and experimental results showing comparable accuracy to exhaustive search across synthetic and benchmark datasets.
Probabilistic time series forecasting remains challenging, largely because modeling distinct trend and seasonal dynamics requires specialized approaches. Existing methods often fail to capture the unique inner properties of these components, lack interpretability, or suffer from heavy memory and runtime overhead.
Watermarking schemes for large language model output are evaluated almost exclusively on English text using each scheme's detection threshold and a narrow set of quality measurements. Multilingual deployment exposes evaluation-design choices that are inconsequential on English but determine conclusions cross-lingually.
Multimodal sarcasm detection aims to identify sarcastic intent from multimodal content, where inconsistencies between literal meaning and contextual cues often signal irony. This task has attracted increasing research attention.
Compositional Zero-Shot Learning (CZSL) aims to recognize unseen attribute-object compositions by leveraging knowledge of primitive concepts learned from seen compositions. Although recent works achieve impressive performance in CZSL by leveraging large vision-language models, they primarily rely on discriminative representations that may not explicitly preserve the structured relationships between primitive concepts and their compositions.
For AI agents to be useful beyond simple chat, they must hold sensitive user context such as calendars, credentials, health records, and financial data. We study whether the mere presence of such secrets in a model's context window introduces hidden correlations into the model's benign outputs, allowing reconstruction even when the model correctly refuses direct extraction.
Buildings account for roughly one-third of global energy consumption and CO$_2$ emissions. Optimizing indoor climate systems plays a critical role for urban climate mitigation aligned with UN Sustainable Development Goals 11 and 13.
LLM evaluation pipelines often have many candidate judges: general LLM-as-a-judge prompts, reward models, safety classifiers, confidence variants, and task-specific verifiers. The deployment question...
TT‑Net introduces a two‑cut tensor‑train decomposition to replace the per‑channel SVD denoising block in conditional GANs, enabling direct cross‑channel information access. In controlled experiments, TT‑Net outperforms SVD‑Net on PSNR and SSIM for Gaussian, motion blur, and salt‑and‑pepper noise, and surpasses EigenGAN and Pix2pix for Gaussian noise. Training dynamics reveal that TT‑Net’s adversarial loss saturates while reconstruction quality continues to improve, raising questions about the role of the adversarial component.
The paper presents a method to align a language‑model‑based crowd agent with aggregate mobility data by fine‑tuning it to match observed destination compositions derived from origin‑to‑destination flows. The approach uses iterative proportional fitting to reweight the model’s destination distribution and corrects for dominant destination inflation by training a low‑rank adapter on resampled trajectories. Experiments on mobile network counts from two baseball games show a 25% reduction in destination‑share error while maintaining similar grid correlation across policies.
Large vision-language models (LVLMs) have achieved remarkable progress in video understanding and reasoning. Despite extensive studies on text- and image-based jailbreaks, video jailbreaks against LVL...