Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Foundations of Stochastic Lexical Calculus: Semantic Descent and Random Dynamics on Probability SimplicesAllen Institute for AI · arXiv (Cornell University)Jul 25OmniScope: Modality-Decoupled Token Compression for Omnimodal Large Language ModelsAlibaba · arXiv (Cornell University)Jul 25Identification and Honest Recovery from Semantic Observation Kernels: Operator Error, Coarsening, and StabilityAllen Institute for AI · arXiv (Cornell University)Jul 25CachedSearch: Training-Free Cached Exploration for Test-Time Search in Video DiffusionAlphabet · arXiv (Cornell University)Jul 25X-Stage: Modeling Post-Issue Backpressure in GPU Communication--Computation FusionNVIDIA · arXiv (Cornell University)Jul 25On the Impossibility of Unbiased and Length-Invariant Policy Optimization with Outcome RewardsAlibaba · arXiv (Cornell University)Jul 25Decoding the Skew: Distribution-Aware MoE Inference with Adaptive Kernel DispatchNVIDIA · arXiv (Cornell University)Jul 25Libra: Taming Attention Workload Skew in Long-Context LLM Training with Bounded Sequence PoolAlibaba · arXiv (Cornell University)Jul 25Capable language models can outgrow the benefits of collaborationAlphabet · Nature Machine Intelligence12 citationsJul 24Coordinated Networking for On-Device Agent-Augmented Real-Time CommunicationMicrosoft · arXiv (Cornell University)Jul 24First-Order In-Context Shapley: Scalable Data Auditing for LLM Alignment and EvaluationAlphabet · arXiv (Cornell University)Jul 24Frustratingly Simple Black-Box Adaptation of Language Models via Logit BiasAlphabet · arXiv (Cornell University)Jul 24Invariant Discovery for Networked SystemsAlphabet · arXiv (Cornell University)Jul 24Learning to Prepare Molecular Ground States with Transformer ModelsNVIDIA · arXiv (Cornell University)Jul 24RadSight: Towards Perceptually Reliable Multimodal Radiology Image UnderstandingAlibaba · arXiv (Cornell University)Jul 24Listen, Do Not Copy: Internalizing Audio-Grounded Scaffold Context for Robust Omni-Model Speech UnderstandingAlibaba · arXiv (Cornell University)Jul 24Are Production Cloud Skills Adequately Tested? Measuring and Governing Skill Test Adequacy in PracticeAlibaba · arXiv (Cornell University)Jul 24TileSight: A First-Principles Tile-Centric Analytical GPU Performance Model from Cores to ClustersMicrosoft · arXiv (Cornell University)Jul 24Where Facts Go Missing: A Layerwise Taxonomy and Per-Layer Attribution of Information Omission in Air-Gapped LLMAgent PipelinesMultiverse Computing · arXiv (Cornell University)Jul 24Introduction to the Special Section on USENIX ATC 2025Alphabet · ACM Transactions on StorageJul 24GeoPandas: Fundamental data structures for vector spatial data in PythonMeta · Computers Environment and Urban SystemsJul 24Site Reliability Engineering Practices for Enhancing Reliability of Large-Scale Cloud ServicesMicrosoft · International Journal of Computer Trends and TechnologyJul 24From classification to optimization: Slicing and resource management with TRACTORNVIDIA · Computer Communications1 citationJul 23Hoare meets Heisenberg: A Lightweight Logic for Quantum ProgramsMicrosoft · Quantum1 citationJul 23Can Generative Recommendation Reach Cold Items? A Temporal Perspective on Semantic-ID GenerationAlibaba · arXiv (Cornell University)Jul 23