Dual-encoder vision-language models systematically fail at negation, scoring ''a cat without a dog'' nearly identically to ''a cat with a dog.'' We show that this failure is inherent to the single-embedding framework. Under contrastive training, single-embedding methods can at best ignore the negated concept (dilution)...
DoYoung Kim, Changkyu Choi, Ad\'in Ram\'irez Rivera et al.· 0 citations
Recent advances in video world models have enabled plausible interactive environments, yet long-horizon generation places a fundamental demand on visual memory: the previously observed world must remain consistent as the model continues to generate. Existing autoregressive approaches often implicitly rely on the genera...
Jun-Yi Wu, Ze-Kun Li, Chenhongyi Yang et al.· 0 citations
Clinical decision-making is a sequential process of information collection, clinical inference, and forecasting under partial observability, often without a uniquely correct treatment decision. Large language models (LLMs) have demonstrated strong medical question-answering capabilities but extending these advances to...
Li-Ching Chen, Andrés Franco Valiente, Yuta Ishiyama et al.· 0 citations
A plug-and-play module for convolutional neural networks that enforces $\mathrm{SE}(2)$ invariance, capturing both translations and rotations within standard architectures, and derive invariant feature maps that are provably optimal: minimal and information-preserving with respect to the symmetry group.
Tomáš Karella, Emily Shinkle, Alice E. A. Allen et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Autonomous quadrotor flight in cluttered environment is an key capability for tasks like inspection and search. Navigation policies trained in simulation and transferred to hardware can fly at high speed in open space, but they can still collide as obstacle clearance shrinks. We attribute most of these collisions to a...
Aditya Azad, Arthur Huang, Guan-Rui Li· 0 citations
Agentic workflows combine large language models with planning, memory, tools, execution environments, and iterative feedback. Their failures are difficult to diagnose because the same user-visible symptom may originate from model behavior, workflow orchestration, external tools, infrastructure, or their interaction. Th...
Vision Transformers incur high computational costs when processing long visual token sequences. Existing reduction methods typically rank tokens independently and discard those deemed unimportant or redundant. We propose a reconstruction-guided framework for visual token condensation (ReConV), which summarizes the comp...
Xiaoya Tang, Shilong Liu, Daniel Smolyak et al.· 0 citations
Generating realistic time-series requires faithfully reproducing rare but high-impact extreme events. Existing approaches often rely on specialized representations or extreme-focused training objectives, which may restrict the diversity of captured extremes or compromise overall distributional fidelity. We propose Cros...
Earthquake catalogs underpin studies such as fault structure, earthquake processes and seismic hazard, and their value depends on being both complete and reliable. Machine learning has improved individual tasks of earthquake monitoring, but processing stations independently discards the consistency of arrivals across t...
Continuous monitoring of generative AI systems requires detecting distribution shifts in the unstructured data they ingest and generate. Existing monitoring methods typically control false alarms by requiring the discrepancy measure used to detect shifts to be *pre-specified*. However, for unstructured data, pre-specif...
Siyu Zhou, Jia-Lin Ouyang, Fan Xia et al.· 1 citation
Day-ahead (DA) electricity price forecasting is usually posed per market, with a separate model fitted to each grid's own history. This caps accuracy: statistical and tree-based baselines saturate, while higher-capacity sequence models need more training data than a single market's few years of hourly prices can supply...
Score-based black-box adversarial attacks pose a serious threat to deployed machine learning systems by exploiting repeated model queries to craft adversarial examples, requiring neither access to model parameters nor architecture. Existing defenses predominantly rely on stochastic output perturbations, which fail to p...
Ali Javeri, Javad Hezareh, Radin Shahdaei et al.· 0 citations
