Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Synthetic Hallucinations, Real Gains: Hard Negatives from Frontier Models for FIM Hallucination MitigationMicrosoft · arXiv (Cornell University)Jun 2The Geometry of LLM-as-Judge: Why Inter-LLM Consensus Is Not Human AlignmentMicrosoft · arXiv (Cornell University)Jun 2Using Reward Uncertainty to Induce Diverse Behaviour in Reinforcement LearningAlphabet · arXiv (Cornell University)Jun 2Reason4Rec: Deliberative User Preference Alignment of Large Language Models for RecommendationMeta · IEEE Transactions on Knowledge and Data EngineeringJun 2Dirichlet Process Mixtures of Block g Priors for Model Selection and Prediction in Linear ModelsAlphabet · Journal of the American Statistical AssociationJun 2Horizontally Scaling Heterogeneous Storage for Combined Capabilities with PolyStoreMicrosoft · ACM Transactions on StorageJun 2From Training-Free to Adaptive: Empirical Insights into MLLMs’ Understanding of Detection InformationAlibaba · International Journal of Computer Vision6 citationsJun 1SPEC CPU: The Next GenerationNVIDIA · Proceedings - International Symposium on Computer Architecture/Proceedings2 citationsJun 1Vistara: Making CXL Real-Full Path From ASIC Design and OS Support to Hyperscale DeploymentMeta · Proceedings - International Symposium on Computer Architecture/Proceedings2 citationsJun 1OpenClaw in the Wild: Security Analysis of Autonomous AgentsAlphabet · IEEE/CAA Journal of Automatica Sinica2 citationsJun 1Loaded Dice: Solving the Non-Selection Problem for Scalable Probabilistic RowHammer DefenseNVIDIA · Proceedings - International Symposium on Computer Architecture/Proceedings1 citationJun 1Progress on the Courtade-Kumar Conjecture: Optimal High-Noise Entropy Bounds and Generalized Coordinate-wise Mutual InformationAlphabet · IEEE International Symposium on Information Theory (ISIT)1 citationJun 1Five-Minute Rule 40 Years Later: A First-Principles Revisit for Modern Memory HierarchyNVIDIA · Proceedings - International Symposium on Computer Architecture/Proceedings1 citationJun 1Context Aware Grounded Teacher for Source Free Object DetectionMicrosoft · International Journal of Computer Vision1 citationJun 1SSEC: Semantic Self-Enhanced Classification for Few-Shot Tabular Data via Adversarial ReasoningAlibaba · International Joint Conference on Neural Networks (IJCNN)Jun 1LM-AdvVC: Localized Multi-Granularity Targeted Adversarial Attacks on Voice ConversioniFlytek · International Joint Conference on Neural Networks (IJCNN)Jun 1OccLLaMA: A Unified Occupancy-Language-Action World Model for Enhancing Motion Planning via Multi-Task LearningHorizon Robotics · Proceedings - IEEE International Conference on Robotics and Automation/ProceedingsJun 1Autonomous Database Tuning Using Machine Learning for Large-Scale Enterprise ApplicationsMeta · Technology and Innovation Conference (TIC)Jun 1Smart Contract-Driven Continuous Auditing: Architecture and EvaluationMicrosoft · Technology and Innovation Conference (TIC)Jun 1Optimizing Distributed SQL Query Execution Through Intelligent Data Partitioning TechniquesMeta · Technology and Innovation Conference (TIC)Jun 17‐1: Real‐Time Neural Filtering for Perceptual Enhancement of Digital ExperiencesNVIDIA · SID Symposium Digest of Technical PapersJun 115‐3: Towards Building Psychophysics‐informed Quality Metrics for Augmented Reality DisplaysMeta · SID Symposium Digest of Technical PapersJun 1Aker: Density-Aware Approximate Caching for Vector SearchMicrosoft · Proceedings of the VLDB EndowmentJun 1dpKernels: Harvesting DPU Compute Resources for Data-Path Efficiency in Cloud Data ProcessingMicrosoft · Proceedings of the VLDB EndowmentJun 1Announcing SIGIR FuturesMicrosoft · ACM SIGIR ForumJun 1