Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
EMO: Pretraining Mixture of Experts for Emergent ModularityAllen Institute for AI · arXiv (Cornell University)May 7Understanding diffusion models requires rethinking (again) generalizationAlphabet · arXiv (Cornell University)May 7One Turn Too Late: Learning When to Intervene Against Multi-Turn Malicious IntentOpenAI · arXiv (Cornell University)May 7DataDignity: Training Data Attribution for Large Language ModelsMicrosoft · arXiv (Cornell University)May 7MAS-Algorithm: A Workflow for Solving Algorithmic Programming Problems with a Multi-Agent SystemAlibaba · arXiv (Cornell University)May 7Three wave mixing vacuum squeezing generation in a SNAIL-based Traveling-Wave Parametric Amplifier with alternated flux polarityAlphabet · arXiv (Cornell University)May 7Beyond Fixed Benchmarks and Worst-Case Attacks: Dynamic Boundary Evaluation for Language ModelsAlphabet · arXiv (Cornell University)May 7Consistent Geometric Deep Learning via Hilbert Bundles and Cellular SheavesSakana AI · arXiv (Cornell University)May 7FreeSpec: Training-Free Long Video Generation via Singular-Spectrum ReconstructionAlibaba · arXiv (Cornell University)May 7AgenticRAG: Agentic Retrieval for Enterprise Knowledge BasesMicrosoft · arXiv (Cornell University)May 7Training Transformers for KV Cache CompressibilityNVIDIA · arXiv (Cornell University)May 7Quantizing With Randomized Hadamard Transforms: Efficient Heuristic Now ProvenMicrosoft · arXiv (Cornell University)May 7SkillOS: Learning Skill Curation for Self-Evolving AgentsAlphabet · arXiv (Cornell University)May 7DASR+: Training Domain Distance Aware Network for Unsupervised Image Super-ResolutionSenseTime · International Journal of Computer VisionMay 7From Reactive to Proactive: Engineering AI-First Reliability for SAP Mission-Critical WorkloadsMicrosoft · International Journal of Research Publications in Engineering Technology and ManagementMay 6SoK: Robustness in Large Language Models against Jailbreak AttacksMicrosoft · arXiv (Cornell University)May 6KEET: Explaining Performance of GPU Kernels Using LLM AgentsNVIDIA · arXiv (Cornell University)May 6SoK: Robustness in Large Language Models against Jailbreak AttacksMicrosoft · arXiv (Cornell University)May 6Quadrature-TreeSHAP: Depth-Independent TreeSHAP and Shapley InteractionsNVIDIA · arXiv (Cornell University)May 6Generative Quantum-inspired Kolmogorov-Arnold EigensolverNVIDIA · arXiv (Cornell University)May 6When KV Meets Embeddings: Dynamic GPU Memory Allocation for Accelerating Generative Recommender ServingAlibaba · arXiv (Cornell University)May 6Tile-Level Activation Overlap for Efficient LLM InferenceMicrosoft · arXiv (Cornell University)May 5GPU-Accelerated Simulations of Moving Boundary Problems and Fluid-Structure Interaction at Extreme ScalesNVIDIA · arXiv (Cornell University)May 5GPU-Accelerated Simulations of Moving Boundary Problems and Fluid-Structure Interaction at Extreme ScalesNVIDIA · arXiv (Cornell University)May 5Learning Dynamics of Zeroth-Order Optimization: A Kernel PerspectiveAlphabet · arXiv (Cornell University)May 5