Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
On the Randomized Locality of Matching Problems in Regular GraphsAlphabet · arXiv (Cornell University)Oct 22, 2025Higher Embedding Dimension Creates a Stronger World Model for a Simple Sorting TaskGoogle DeepMind · arXiv (Cornell University)Oct 21, 2025ACTG-ARL: Differentially Private Conditional Text Generation with RL-Boosted ControlAlphabet · arXiv (Cornell University)Oct 21, 2025Build Stable TrianglesAlphabet · IEEE SoftwareOct 21, 2025The discrete adiabatic quantum linear system solver has lower constant factors than the randomized adiabatic solverAlphabet · Quantum7 citationsOct 20, 2025Bolt3D: Generating 3D Scenes in SecondsAlphabet, Google DeepMind · IEEE/CVF International Conference on Computer Vision (ICCV)11 citationsOct 19, 2025StochasticSplats: Stochastic Rasterization for Sorting-Free 3D Gaussian SplattingAlphabet, Google DeepMind · IEEE/CVF International Conference on Computer Vision (ICCV)9 citationsOct 19, 2025EVER: Exact Volumetric Ellipsoid Rendering for Real-Time View SynthesisAlphabet · IEEE/CVF International Conference on Computer Vision (ICCV)9 citationsOct 19, 2025Tapnext: Tracking Any Point (Tap) as Next Token PredictionGoogle DeepMind · IEEE/CVF International Conference on Computer Vision (ICCV)8 citationsOct 19, 2025Minerva: Evaluating Complex Video ReasoningGoogle DeepMind · IEEE/CVF International Conference on Computer Vision (ICCV)7 citationsOct 19, 2025Erasing More Than Intended? How Concept Erasure Degrades the Generation of Non-Target ConceptsAlphabet, Google DeepMind · IEEE/CVF International Conference on Computer Vision (ICCV)4 citationsOct 19, 2025RoMo: Robust Motion Segmentation Improves Structure from MotionGoogle DeepMind · IEEE/CVF International Conference on Computer Vision (ICCV)4 citationsOct 19, 2025Towards Cross-Modal Backward-Compatible Representation Learning for Vision-Language ModelsGoogle DeepMind · IEEE/CVF International Conference on Computer Vision (ICCV)4 citationsOct 19, 2025ObjectMate: A Recurrence Prior for Object Insertion and Subject-Driven GenerationAlphabet · IEEE/CVF International Conference on Computer Vision (ICCV)3 citationsOct 19, 2025MFT-VITON: High-Fidelity Virtual Try-On with Minimal Input via a Mask-Free Transformer-Diffusion ModelAlphabet · IEEE International Conference on Computer Vision Workshops2 citationsOct 19, 2025Spectral Image TokenizerAlphabet · IEEE/CVF International Conference on Computer Vision (ICCV)2 citationsOct 19, 2025Consistent Time-of-Flight Depth Denoising via Graph-Informed Geometric AttentionAlphabet · IEEE/CVF International Conference on Computer Vision (ICCV)1 citationOct 19, 2025TAB: Transformer Attention Bottlenecks Enable User Intervention and Debugging in Vision-Language ModelsGoogle DeepMind · IEEE/CVF International Conference on Computer Vision (ICCV)1 citationOct 19, 2025Visual Chronicles: Using Multimodal LLMs to Analyze Massive Collections of ImagesGoogle DeepMind · IEEE/CVF International Conference on Computer Vision (ICCV)1 citationOct 19, 2025Infusing Fine-Grained Visual Knowledge to Vision-Language ModelsGoogle DeepMind · IEEE International Conference on Computer Vision Workshops1 citationOct 19, 2025Global-to-Local or Local-to-Global? Enhancing Image Retrieval with Efficient Local Search and Effective Global Re-RankingAlphabet, Google DeepMind · IEEE International Conference on Computer Vision Workshops1 citationOct 19, 2025Principles of Visual Tokens for Efficient Video UnderstandingGoogle DeepMind · IEEE/CVF International Conference on Computer Vision (ICCV)1 citationOct 19, 2025Magic Insert: Style-Aware Drag-And-DropAlphabet, Google DeepMind · IEEE/CVF International Conference on Computer Vision (ICCV)1 citationOct 19, 2025LayerLock: Non-Collapsing Representation Learning with Progressive FreezingGoogle DeepMind · IEEE/CVF International Conference on Computer Vision (ICCV)Oct 19, 2025From Prompt to Progression: Taming Video Diffusion Models for Seamless Attribute TransitionGoogle DeepMind · IEEE/CVF International Conference on Computer Vision (ICCV)Oct 19, 2025