Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
When Models Leave the Training Distribution: A Tutorial on OOD DetectionAlphabet · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)Aug 6RelKD 2026: The Fourth International Workshop on Resource-Efficient Learning for Knowledge DiscoveryAlphabet · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)Aug 6Vectorizing the Trie: Efficient Constrained Decoding for LLM-based Generative Retrieval on AcceleratorsAlphabet · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)Aug 6Deconstructing Staleness: Mitigating Training-Serving Skew in Industrial-scale Online Learning ModelsAlphabet · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)Aug 6Addressing Intent Dissonance in Cross-Domain Distillation on YouTubeAlphabet · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)Aug 6FedKDD/FedMAS 2026: The 2026 International Joint Workshop on Federated Learning for Multi-agent Systems and Data MiningAlphabet · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)Aug 6A UNIQUE REPRESENTATION OF RANKINGS ON PARTIAL DERIVATIVESAlphabet · Journal of Mathematical SciencesAug 6SSTQ:Privacy-Preserving Vector Quantization via Subsampled Stochastic TurboQuantAlphabet · arXiv (Cornell University)Aug 5MoEGen: Mixture-of-Experts for Instance-Adaptive LoRA GenerationAlphabet · arXiv (Cornell University)Aug 4RUTA: Principled Visual Token Allocation via Rate-Utility OptimizationAlphabet · arXiv (Cornell University)Aug 4TimeRLM: Recursive Language Models Enable Precise Anomaly Localization in Long-Context Time-SeriesAlphabet · arXiv (Cornell University)Aug 4TaskPress: Query-Agnostic KV Cache Compression via Task-Guided PruningAlphabet · arXiv (Cornell University)Aug 4Pairwise-Independent Dithering for Single-Stage Hadamard QuantizationAlphabet · arXiv (Cornell University)Aug 3On Topology's Role in ML Training PerformanceAlphabet · arXiv (Cornell University)Aug 3EduZone: A Framework for Evaluating LLM Safety for K-12 Students and TeachersAlphabet · arXiv (Cornell University)Aug 3The Condition-Number Barrier in Sparse Least SquaresAlphabet · arXiv (Cornell University)Aug 3Paras: Actively Secure Two-Server Private HistogramsAlphabet · IACR Communications in CryptologyAug 3HueManity: Probing Fine-Grained Visual Perception in MLLMsAlphabet · Lecture notes in computer scienceAug 2Bridging Algorithmic Information Theory and Machine Learning, Part IV: Solomonoff Gaussian Hilbert Spaces, Solomonoff Gaussian Processes, and Solomonoff Gaussian FieldsGoogle DeepMind · Physica D Nonlinear Phenomena2 citationsAug 1Effective Context Management for LLM Agents via Hybrid Chat History OptimizationAlphabet · IEEE World Conference on Applied Intelligence and Computing (AIC)Aug 1MAAT: Applied Computational Intelligence for Autonomous Software Testing via Deep Learning, Reinforcement Learning and Knowledge-Based ReasoningAlphabet · IEEE World Conference on Applied Intelligence and Computing (AIC)Aug 1Finding the Real Bottleneck in Large-Scale ML TrainingAlphabet · QueueAug 1Grey Wolf Optimization Method Based on Dynamic Workload Allocation for Video Conferencing SystemAlphabet, Microsoft · International Conference Computational Vision and Bio Inspired ComputingAug 1iPDB: SQL with ML and LLM Predicates (Towards a Database Engine for AI)Alphabet · Proceedings of the VLDB EndowmentAug 1The Live Database: Firestore's Scalable and Consistent Realtime QueriesAlphabet · Proceedings of the VLDB EndowmentAug 1