Google says its new TurboQuant method could improve how efficiently AI models run by compressing the key-value cache used in LLM inference and supporting more efficient vector search.
In tests on Gemma and Mistral models, the …
arXiv:2509.22460v3 Announce Type: replace
Abstract: Geometric Problem Solving (GPS) poses a unique challenge for Multimodal Large Language Models (MLLMs), requiring not only the joint interpretation of text and diagrams but also iterative visuospatial…
arXiv:2603.22593v2 Announce Type: replace-cross
Abstract: Sparse Autoencoders uncover thousands of features in vision models, yet explaining these features without requiring human intervention remains an open challenge. While previous work has propose…
arXiv:2603.14185v3 Announce Type: replace
Abstract: Generative multimodal models can exhibit safety failures that are inherently relational: two benign concepts can become unsafe when linked by a specific action or relation (e.g., child-drinking-wine)…
arXiv:2603.24584v1 Announce Type: new
Abstract: Vision–Language–Action (VLA) policies have shown strong progress in mapping language instructions and visual observations to robotic actions, yet their reliability degrades in cluttered scenes with dis…
arXiv:2511.13649v4 Announce Type: replace
Abstract: Distribution Matching Distillation (DMD) facilitates efficient inference by distilling multi-step diffusion models into few-step variants. Concurrently, Reinforcement Learning (RL) has emerged as a v…
arXiv:2603.23845v1 Announce Type: new
Abstract: Deep learning and generative models are advancing rapidly, with synthetic data increasingly being integrated into training pipelines for downstream analysis tasks. However, in medical imaging, their adop…
arXiv:2512.18128v3 Announce Type: replace
Abstract: High-resolution mapping of canopy height is essential for forest management and biodiversity monitoring. Although recent studies have led to the advent of deep learning methods using satellite imager…
arXiv:2603.23868v1 Announce Type: new
Abstract: In this paper, we address the challenging problem of single-scene, fully unsupervised video anomaly detection (VAD), where raw videos containing both normal and abnormal events are used directly for trai…
arXiv:2603.01601v2 Announce Type: replace
Abstract: Large 3D reconstruction models have revolutionized the 3D content generation field, enabling broad applications in virtual reality and gaming. Just like other large models, large 3D reconstruction mo…