Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Train Separately, Merge Together: Modular Post-Training with Mixture-of-ExpertsAllen Institute for AI · arXiv (Cornell University)Apr 20Revisiting Code Debloating with Ground Truth-based EvaluationAlphabet · arXiv (Cornell University)Apr 20Modeling Multiple Support Strategies within a Single Turn for Emotional Support ConversationsAlibaba · arXiv (Cornell University)Apr 20Sonata: A Hybrid World Model for Inertial Kinematics under Clinical Data ScarcityNVIDIA · arXiv (Cornell University)Apr 20Video-Based Arabic Sign Language Recognition with Mediapipe and Deep Learning TechniquesAlphabet · Journal of ImagingApr 20Post-quantum cryptography for space systems: Algorithms, implementation, and design constraints—A systematic surveyTechnology Innovation Institute · Acta AstronauticaApr 20Confounder Balance in Next Basket PredictionNVIDIA · IEEE Transactions on CyberneticsApr 20Q-Regularized Generative Auto-Bidding: From Suboptimal Trajectories to Optimal PoliciesAlibaba · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)Apr 20Macro Graph of Experts for Billion-Scale Multi-Task RecommendationAlibaba · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)Apr 20Meta Lattice: Model Space Redesign for Cost-Effective Industry-Scale Ads RecommendationsMeta · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)Apr 20ALF: Advertiser Large Foundation Model for Multi-Modal Advertiser UnderstandingAlphabet · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)Apr 20Sell It Before You Make It: Revolutionizing E-Commerce with Personalized AI-Generated ItemsAlibaba · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)Apr 20Transparent and Controllable Recommendation Filtering via Multimodal Multi-Agent CollaborationMicrosoft · arXiv (Cornell University)1 citationApr 19A 184mW, 61fps Image Signal Processor for Spatial-Temporal HD Video InpaintingAlphabet · Custom Integrated Circuits ConferenceApr 19HorizonBench: Long-Horizon Personalization with Evolving PreferencesOpenAI · arXiv (Cornell University)Apr 19Flint: Compiler Enabled Cluster-Free Design Space Exploration for Distributed MLNVIDIA · arXiv (Cornell University)Apr 19Waking Up Blind: Cold-Start Optimization of Supervision-Free Agentic Trajectories for Grounded Visual PerceptionMicrosoft · arXiv (Cornell University)Apr 19MoVE: Translating Laughter and Tears via Mixture of Vocalization Experts in Speech-to-Speech TranslationNVIDIA · arXiv (Cornell University)Apr 19HORIZON: A Benchmark for In-the-wild User Behaviour ModelingMicrosoft · arXiv (Cornell University)Apr 19SpatialImaginer: Towards Adaptive Visual Imagination for Spatial ReasoningMiniMax · arXiv (Cornell University)Apr 19TimeRFT: Stimulating Generalizable Time Series Forecasting for TSFMs via Reinforcement FinetuningAlibaba · arXiv (Cornell University)Apr 18Adaptive Forensic Feature Refinement via Intrinsic Importance PerceptionAlibaba · arXiv (Cornell University)Apr 18SeekerGym: A Benchmark for Reliable Information SeekingAlphabet · arXiv (Cornell University)Apr 18Incentivizing Parametric Knowledge via Reinforcement Learning with Verifiable Rewards for Cross-Cultural Entity TranslationAlibaba · arXiv (Cornell University)Apr 18AutoPKG: An Automated Framework for Dynamic E-commerce Product-Attribute Knowledge Graph ConstructionAlibaba · arXiv (Cornell University)Apr 18