Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
How Does Alignment Enhance LLMs' Multilingual Capabilities? A Language Neurons PerspectiveMicrosoft · arXiv (Cornell University)5 citationsMay 27, 2025Universal and Tight Bounds on Counting Errors of Count-Min Sketch with Conservative UpdatesMeta · Proceedings of the ACM on Measurement and Analysis of Computing Systems5 citationsMay 27, 2025Revisiting Traffic Splitting for Software Switch in DatacenterMicrosoft · Proceedings of the ACM on Measurement and Analysis of Computing Systems2 citationsMay 27, 2025InstGenIE: Generative Image Editing Made Efficient with Mask-aware Caching and SchedulingAlibaba · arXiv (Cornell University)1 citationMay 27, 2025Integrating Experimental, Numerical and Machine Learning Models for Real-Time, Efficient Data Center Cooling ControlNVIDIA · I-THERM :/I-THERM - Intersociety Conference on Thermal Phenomena in Electronic SystemsMay 27, 2025Frequency Composition for Compressed and Domain-Adaptive Neural NetworksAlphabet · arXiv (Cornell University)May 27, 2025Auto-tables: synthesizing multi-step transformations to relationalize tables without using examplesMicrosoft · The VLDB JournalMay 27, 2025Scaling and logic in the colour code on a superconducting quantum processorAlphabet, Google DeepMind · Nature91 citationsMay 26, 2025SafeDPO: A Simple Approach to Direct Preference Optimization with Enhanced SafetyLG AI Research · arXiv (Cornell University)32 citationsMay 26, 2025Token-Importance Guided Direct Preference OptimizationMicrosoft · arXiv (Cornell University)26 citationsMay 26, 2025MaskSearch: A Universal Pre-Training Framework to Enhance Agentic Search CapabilityAlibaba · arXiv (Cornell University)23 citationsMay 26, 2025Beyond Markovian: Reflective Exploration via Bayes-Adaptive RL for LLM ReasoningAlphabet, Google DeepMind · arXiv (Cornell University)12 citationsMay 26, 2025Proximal Point Nash Learning from Human FeedbackGoogle DeepMind, Hugging Face · arXiv (Cornell University)12 citationsMay 26, 2025LLaFS++: Few-Shot Image Segmentation With Large Language ModelsAlibaba · IEEE Transactions on Pattern Analysis and Machine Intelligence9 citationsMay 26, 2025Characterizing Pattern Matching and Its Limits on Compositional Task StructuresLG AI Research · arXiv (Cornell University)6 citationsMay 26, 2025Beyond Safe Answers: A Benchmark for Evaluating True Risk Awareness in Large Reasoning ModelsAlibaba · arXiv (Cornell University)6 citationsMay 26, 2025Are Time-Series Foundation Models Deployment-Ready? A Systematic Study of Adversarial Robustness Across DomainsMicrosoft · arXiv (Cornell University)6 citationsMay 26, 2025BASE-Q: Bias and Asymmetric Scaling Enhanced Rotational Quantization for Large Language ModelsAlibaba · arXiv (Cornell University)5 citationsMay 26, 2025AutoViT: Achieving Real-Time Vision Transformers on Mobile via Latency-aware Coarse-to-Fine SearchMicrosoft · International Journal of Computer Vision5 citationsMay 26, 2025Zero-Shot Streaming Text to Speech Synthesis with Transducer and Auto-Regressive ModelingMicrosoft · arXiv (Cornell University)4 citationsMay 26, 2025VSCBench: Bridging the Gap in Vision-Language Model Safety CalibrationAlibaba · arXiv (Cornell University)3 citationsMay 26, 2025Decomposing Complex Visual Comprehension into Atomic Visual Skills for Vision Language ModelsGoogle DeepMind · arXiv (Cornell University)3 citationsMay 26, 2025The Limits of Preference Data for Post-TrainingAlphabet · arXiv (Cornell University)3 citationsMay 26, 2025TESSER: Transfer-Enhancing Adversarial Attacks from Vision Transformers via Spectral and Semantic RegularizationTechnology Innovation Institute · arXiv (Cornell University)2 citationsMay 26, 2025Boosting Time-Series Domain Adaptation via A Time-Frequency Consensus FrameworkTechnology Innovation Institute · IEEE Transactions on Artificial Intelligence2 citationsMay 26, 2025