Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Performance Aware LLM Load Balancer for Mixed WorkloadsMicrosoft · Workshop on Machine Learning and Systems (EuroMLSys)17 citationsMar 30, 2025Re-Aligning Language to Visual Objects with an Agentic WorkflowSenseTime · arXiv (Cornell University)3 citationsMar 30, 2025The Integration of NLP and Computer Vision: Advanced Frameworks for Multi-Modal Content UnderstandingMeta · International Journal of Scientific Research in Computer Science Engineering and Information Technology1 citationMar 30, 2025SpotLessSplats: Ignoring Distractors in 3D Gaussian SplattingAlphabet · ACM Transactions on Graphics83 citationsMar 29, 2025UniHDSA: A unified relation prediction approach for hierarchical document structure analysisMicrosoft · Pattern Recognition7 citationsMar 29, 2025Evaluating Compositional Scene Understanding in Multimodal Generative ModelsMicrosoft · arXiv (Cornell University)5 citationsMar 29, 2025Factored Agents: Decoupling In-Context Learning and Memorization for Robust Tool UseAlphabet · arXiv (Cornell University)4 citationsMar 29, 2025The Earth alignment principle for artificial intelligenceMicrosoft · Nature Sustainability17 citationsMar 28, 2025Niyama : Breaking the Silos of LLM Inference ServingMicrosoft · arXiv (Cornell University)13 citationsMar 28, 2025A Framework for Large-Scale Synthetic Graph Dataset GenerationNVIDIA · IEEE Transactions on Neural Networks and Learning Systems11 citationsMar 28, 2025Task Tokens: A Flexible Approach to Adapting Behavior Foundation ModelsNVIDIA · arXiv (Cornell University)4 citationsMar 28, 2025AI-Driven Optimization of Cloud Data Centers: Enhancing Reliability and Cost EfficiencyAlphabet · International Conference on Data Science, Agents & Artificial Intelligence (ICDSAAI)3 citationsMar 28, 2025Text2Bundle: Towards Personalized Query-based Bundle GenerationiFlytek · ACM Transactions on Recommender Systems3 citationsMar 28, 2025Exploiting spatial information and target speaker phoneme loss for multichannel directional speech enhancement and recognitionAlibaba · Computer Speech & Language1 citationMar 28, 2025PIM Is All You Need: A CXL-Enabled GPU-Free System for Large Language Model InferenceAlphabet · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)68 citationsMar 27, 2025Systematic CXL Memory Characterization and Performance Analysis at ScaleMicrosoft · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)66 citationsMar 27, 2025TAPAS: Thermal- and Power-Aware Scheduling for LLM Inference in Cloud PlatformsMicrosoft · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)65 citationsMar 27, 2025POD-A ttention : Unlocking Full Prefill-Decode Overlap for Faster LLM InferenceMicrosoft · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)60 citationsMar 27, 2025All for one, and one for all: UrbanSyn Dataset, the third Musketeer of synthetic driving scenesNVIDIA · Neurocomputing43 citationsMar 27, 2025Relax: Composable Abstractions for End-to-End Dynamic Machine LearningNVIDIA, OpenAI · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)30 citationsMar 27, 2025debug-gym: A Text-Based Environment for Interactive DebuggingMicrosoft · arXiv (Cornell University)29 citationsMar 27, 2025Shared Global and Local Geometry of Language Model EmbeddingsGoogle DeepMind · arXiv (Cornell University)28 citationsMar 27, 2025Bi-VLDoc: bidirectional vision-language modeling for visually-rich document understandingAlibaba · International Journal on Document Analysis and Recognition (IJDAR)22 citationsMar 27, 2025Past-Future Scheduler for LLM Serving under SLA GuaranteesSenseTime · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)22 citationsMar 27, 2025Hardware Sentinel: Protecting Software Applications from Hardware Silent Data CorruptionsMeta · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)13 citationsMar 27, 2025