Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Marginalized Bundle Adjustment: Multi-View Camera Pose from Monocular Depth EstimatesAlphabet · International Conference on 3D Vision (3DV)Mar 20"It didn't feel right but I needed a job so desperately": Understanding People's Emotions & Help Needs During Financial ScamsAlphabet · arXiv (Cornell University)Mar 20Personalization at Scale: AI-Driven User Experience OptimizationMeta · Journal of Artificial Intelligence & Cloud ComputingMar 20AgenticRS-EnsNAS: Ensemble-Decoupled Self-Evolving Architecture SearchAlibaba · arXiv (Cornell University)Mar 20AIGQ: An End-to-End Hybrid Generative Architecture for E-commerce Query RecommendationAlibaba · arXiv (Cornell University)Mar 20Beyond Single Tokens: Distilling Discrete Diffusion Models via Discrete MMDAlphabet · arXiv (Cornell University)Mar 20An Agentic Multi-Agent Architecture for Cybersecurity Risk ManagementMicrosoft · arXiv (Cornell University)Mar 20Demonstrations, CoT, and Prompting: A Theoretical Analysis of ICLMicrosoft · arXiv (Cornell University)Mar 20PoC: Performance-oriented Context Compression for Large Language Models via Performance PredictionAlibaba · arXiv (Cornell University)Mar 20Neither Here Nor There: Cross-Lingual Representation Dynamics of Code-Mixed Text in Multilingual EncodersMicrosoft · arXiv (Cornell University)Mar 20LLM-Enhanced Failure Localization in Microservices: Integrating Multi-Modal Data and Expert InterpretationMicrosoft · IEEE Transactions on Services ComputingMar 20Enhancing 3D medical multi-modal large language models with integrated human body priors for computed tomographyAlibaba · Pattern Recognition1 citationMar 19Secure and Privacy-Preserving Healthcare Federated Learning via Differential Privacy MechanismsAlphabet · International Symposium on Digital Forensics and Security (ISDFS)Mar 19How Auditory Knowledge in LLM Backbones Shapes Audio Language Models: A Holistic EvaluationNVIDIA · arXiv (Cornell University)Mar 19Seasoning Generative Models for a Generalization AftertasteAlphabet, Google DeepMind · arXiv (Cornell University)Mar 19Parallelizing Tool Execution and LLM Generation for Low-Latency Agent ServingMicrosoft · arXiv (Cornell University)Mar 19Counting Circuits: Mechanistic Interpretability of Visual Reasoning in Large Vision-Language ModelsAllen Institute for AI · arXiv (Cornell University)Mar 19From Servers to Sites: Compositional Power Trace Generation of LLM Inference for Infrastructure PlanningMicrosoft · arXiv (Cornell University)Mar 19Rethinking Retrieval-Augmentation as Synthesis: A Query-Aware Context Merging ApproachAlibaba · arXiv (Cornell University)Mar 18Learning to Reason with Curriculum I: Provable Benefits of AutocurriculumMicrosoft · arXiv (Cornell University)Mar 18Beyond Forced Modality Balance: Intrinsic Information Budgets for Multimodal LearningAlibaba · arXiv (Cornell University)Mar 18Frozen-State Value Iteration: Faster Reinforcement Learning by Freezing Slow StatesMeta · Management ScienceMar 18Inhibitory normalization of error signals improves learning in neural circuitsAlphabet · arXiv (Cornell University)Mar 18Revisiting Cross-Attention Mechanisms: Leveraging Beneficial Noise for Domain-Adaptive LearningAlibaba · arXiv (Cornell University)Mar 18Weight Clustering Reveals the Importance of Relative Rank in Large Language ModelsMultiverse Computing · arXiv (Cornell University)Mar 18