Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Meta-Initialized Warm Pool Manager: Predictive Cold Start Mitigation for Serverless Functions via Adaptive Meta-LearningAlphabet · IEEE International Conference for Convergence in Computing Technology (I3CTCON)Mar 14Missing Mass for Differentially Private Domain DiscoveryAlphabet · arXiv (Cornell University)Mar 14Semantic Document Derendering: SVG Reconstruction via Vision-Language ModelingGoogle DeepMind · Proceedings of the AAAI Conference on Artificial IntelligenceMar 14Improved Differentially Private Algorithms for Rank AggregationAlphabet · Proceedings of the AAAI Conference on Artificial IntelligenceMar 14Policy Newton Methods for Distortion RiskmetricsAlphabet · Proceedings of the AAAI Conference on Artificial IntelligenceMar 14Beyond Fixed Tasks: Unsupervised Environment Design for Task-Level PairsGoogle DeepMind · Proceedings of the AAAI Conference on Artificial IntelligenceMar 14VirtualEnv: A Platform for Embodied AI ResearchAlphabet, Google DeepMind · Proceedings of the AAAI Conference on Artificial IntelligenceMar 14Enhancing Search Efficiency through LLM-Based User Memory Systems for Query Matching and Intent ModelingAlphabet · International Conference on Cyber Security, Artificial Intelligence and Digital Economy (CSAIDE)Mar 13When Deep Learning Meets Polyhedral Theory: A SurveyAlphabet · INFORMS journal on computing20 citationsMar 12Measurement-Induced State transitions in Inductively-Shunted TransmonsAlphabet · arXiv (Cornell University)Mar 12Taking Shortcuts for Categorical VQA Using Super NeuronsAlphabet, Google DeepMind · arXiv (Cornell University)Mar 11What do near-optimal learning rate schedules look like?Google DeepMind · arXiv (Cornell University)Mar 11Diagnostic accuracy, fairness and clinical implementation of AI for breast cancer screening: results of multicenter retrospective and prospective technical feasibility studiesAlphabet · Nature Cancer16 citationsMar 10Impact of using artificial intelligence as a second reader in breast screening including arbitrationAlphabet · Nature Cancer5 citationsMar 10Expert evaluation of LLM world models: A high-T c superconductivity case studyAlphabet · Proceedings of the National Academy of Sciences2 citationsMar 10Triton-Sanitizer: A Fast and Device-Agnostic Memory Sanitizer for Triton with Rich Diagnostic ContextAlphabet, Anthropic, Meta, OpenAI · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)1 citationMar 10Optimal partition selection with Rényi differential privacyAlphabet · arXiv (Cornell University)Mar 10Denoising the US Census: Succinct Block Hierarchical RegressionAlphabet · arXiv (Cornell University)Mar 10GS-Scale: Unlocking Large-Scale 3D Gaussian Splatting Training via Host OffloadingAlphabet · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)Mar 10Optimal partition selection with Rényi differential privacyAlphabet · arXiv (Cornell University)Mar 10Neura: A Unified Framework for Hierarchical and Adaptive CGRAsAlphabet · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)Mar 10CLIOPATRA: Extracting Private Information from LLM InsightsAlphabet · arXiv (Cornell University)Mar 10Thinking to Recall: How Reasoning Unlocks Parametric Knowledge in LLMsAlphabet · arXiv (Cornell University)Mar 10SpecProto: A Parallelizing Compiler for Speculative Decoding of Large Protocol Buffers DataAlphabet · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)Mar 10An experimental evaluation of an AI-powered interactive learning platformAlphabet · Frontiers in Artificial IntelligenceMar 10