Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
RRSI: Regularized Recursive Self-Improvement of Agent HarnessesAlphabet · arXiv (Cornell University)Sep 21Why Ghost Outputs Teach: A Kernel-Based Understanding of Subliminal LearningAlphabet · arXiv (Cornell University)Sep 20STEVE: Stabilizing Textual Gradient-Based Prompt Optimization via Error-Driven Refinement and Regularized VerificationGoogle DeepMind · arXiv (Cornell University)Sep 20Explainable Recommendations at Scale: LLM Rationales for YouTube Music Artist DiscoveryAlphabet · arXiv (Cornell University)Sep 20Extracting Electromagnetic Bare Mode Couplings in Large Superconducting Quantum ProcessorsAlphabet · arXiv (Cornell University)Sep 18Pretrained Persona Mixture Models and Tandem Models for Human SimulationAlphabet · arXiv (Cornell University)Sep 18Presage: Prefetch Search via Agent-Guided ExperimentsAlphabet · arXiv (Cornell University)Sep 18Denser Planar Color CodesAlphabet · arXiv (Cornell University)Sep 18Geometric Mean Pooling for Equal-Weight Multiplicative Coarse-GrainingAlphabet · arXiv (Cornell University)Sep 18When Does Reasoning Help in Machine Translation? A Hierarchical Analysis of LRM Reasoning TracesGoogle DeepMind · arXiv (Cornell University)Sep 18Trust, but Validate the Instrument: Auditing AI-Generated RTL Verification Plans on Authored Security-Regression ProxiesAlphabet · arXiv (Cornell University)Sep 17The Stochastic Shift: A New Evaluation Paradigm for Text-to-SQL with AI OperatorsAlphabet · arXiv (Cornell University)Sep 17DLB: Distributed Load Balancing at Scale for Generative AI InferenceAlphabet · arXiv (Cornell University)Sep 17Harnessing Generative UI for Education: Tailored Learning InteractivesAlphabet · arXiv (Cornell University)Sep 17FunPilot: Runtime Performance Diagnosis and Remediation for Serverless Applications with LLMsAlphabet · ACM SIGOPS Asia-Pacific Workshop on Systems (APSys)Sep 17denser planar color codesAlphabet · arXiv (Cornell University)Sep 17DualSQL: Text-to-SQL with Multi-Agent Reinforcement LearningAlphabet · arXiv (Cornell University)Sep 16Decodability is Not Causality: Dissociating Probe Readouts from Behavioral Drivers via SAE DecompositionAlphabet · arXiv (Cornell University)Sep 16Locating Hidden Failures Makes Long-Horizon Agents More ReliableAlphabet, Google DeepMind · arXiv (Cornell University)Sep 15Tight Lower Bounds for Differentially Private Continual CountingAlphabet · arXiv (Cornell University)Sep 15Query-Aware Source-Risk Triage for Retrieval-Augmented GenerationAlphabet · arXiv (Cornell University)Sep 15Verifiable Social Reasoning for LLM AssistantsAlphabet · arXiv (Cornell University)Sep 15High Probability Streaming Lower Bounds for $F_2$ EstimationAlphabet · arXiv (Cornell University)Sep 15RadarMind: cognitive AI with mmWave sensing for human-aware intelligent automationAlphabet, Microsoft · Journal on Advances in Signal ProcessingSep 15Towards Scalable Measurement of Durable SkillsAlphabet · arXiv (Cornell University)Sep 14