Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Co-Harness: Co-Evolving Harnesses and Model Weights for LLM AgentsAllen Institute for AI · arXiv (Cornell University)Jul 17Scalable LLM Agent Tool Access in the CloudAlibaba · arXiv (Cornell University)Jul 17Audio-Visual Flamingo: Open Audio-Visual Intelligence for Long and Complex VideosNVIDIA · arXiv (Cornell University)Jul 17QUADS: Stabilizing NVFP4 Reinforcement Learning for MoE via QUantization-error Alignment across Dual SidesAlibaba · arXiv (Cornell University)Jul 17Adaptive Fault Injection Planning for Multi-Layer Self-Healing AI InfrastructureMeta · arXiv (Cornell University)Jul 17Every Microsecond Matters: Achieving Near Speed-of-Light Latency in GPU CollectivesNVIDIA · arXiv (Cornell University)Jul 17Before the Action: Benchmarking LLMs on Prospective Hypothesis DiscoveryAlibaba · arXiv (Cornell University)Jul 17Accelerating FrodoKEM in HardwareMicrosoft · IACR Transactions on Cryptographic Hardware and Embedded SystemsJul 17Enhancing Offline Model-Based RL via Active Model Selection: A Bayesian Optimization PerspectiveMeta · Machine LearningJul 17MACE-Dance: Motion-Appearance Cascaded Experts for Music-Driven Dance Video GenerationAlibaba · ACM SIGGRAPH Conference Papers (SIGGRAPH)3 citationsJul 16Performance Model for Hybrid Quantum-Classical WorkflowsNVIDIA · arXiv (Cornell University)Jul 16Spectral amplification for ground-state energy estimation of electronic structure in first quantizationAlphabet · arXiv (Cornell University)Jul 16Are LLM-Generated GPU Kernels Production-Ready? A Trace-Driven Benchmark and Optimization AgentAlibaba · arXiv (Cornell University)Jul 16CODA: Algorithm-Hardware Co-design for Edge Video Diffusion via NMP-Enabled Compute-Cache Operator DisaggregationAlibaba · arXiv (Cornell University)Jul 16Spectral Dual Fitting for $k$-MeansAlphabet · arXiv (Cornell University)Jul 16AlphaWiSE: Adaptive Weight Interpolation for Continual Multimodal Representation LearningGoogle DeepMind · arXiv (Cornell University)Jul 16Pretraining Data Can Be Poisoned through Computational PropagandaAllen Institute for AI · arXiv (Cornell University)Jul 16TSMC: Time-varying 4D Scene Mesh CompressionNVIDIA · ACM SIGGRAPH Conference Papers (SIGGRAPH)Jul 16Toward Richer Material Generation via Procedural Data EnhancementNVIDIA · ACM SIGGRAPH Conference Papers (SIGGRAPH)Jul 168DNA: 8D Neural Asset Light Transport by Distribution LearningNVIDIA · ACM SIGGRAPH Conference Papers (SIGGRAPH)Jul 16Spatio-Temporal Control Variates with ReSTIR for Real-Time RenderingNVIDIA · ACM SIGGRAPH Conference Papers (SIGGRAPH)Jul 16VideoNeuMat: Neural Material Extraction from Generative Video ModelsNVIDIA · ACM SIGGRAPH Conference Papers (SIGGRAPH)Jul 16Adapting Quality Metrics to Tone MappingMeta · ACM SIGGRAPH Conference Papers (SIGGRAPH)Jul 16Divide and Truncate: A Penetration and Inversion Free Framework for Coupled Multi-physics SystemNVIDIA · ACM SIGGRAPH Conference Papers (SIGGRAPH)Jul 16UCM: Unified Modeling of Camera Control and Memory with Time-aware Positional Encoding Warping for World ModelsAlibaba · ACM SIGGRAPH Conference Papers (SIGGRAPH)Jul 16