Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Neural FOXP2 -- Language Specific Neuron Steering for Targeted Language Improvement in LLMsMeta · arXiv (Cornell University)Feb 1On the Expressive Power of Permutation-Equivariant Weight-Space NetworksNVIDIA · arXiv (Cornell University)Feb 1The Horizon Threshold in Cooperative Multi-Agent Reward-Free ExplorationAlphabet · arXiv (Cornell University)Feb 1PromptRL: Prompt Matters in RL for Flow-Based Image GenerationMeta · arXiv (Cornell University)Feb 1ReLayout: Versatile and Structure-Preserving Design Layout Editing via Relation-Aware Design ReconstructionMicrosoft · arXiv (Cornell University)Feb 1From Pragmas to Partners: A Symbiotic Evolution of Agentic High-Level SynthesisCerebras Systems · arXiv (Cornell University)Feb 1Too Good to Be Human: AI Turing Test in Incomplete Information-Competitive GamesAlibaba · ComputerFeb 1What Does Vision Tool-Use Reinforcement Learning Really Learn? Disentangling Tool-Induced and Intrinsic Effects for Crop-and-ZoomMiniMax · arXiv (Cornell University)Feb 1Your Self-Play Algorithm is Secretly an Adversarial Imitator: Understanding LLM Self-Play through the Lens of Imitation LearningMicrosoft · arXiv (Cornell University)Feb 1Flow matching meets biology and life science: a surveyMeta · npj Artificial Intelligence12 citationsJan 31Leveraging ASIC AI Chips for Homomorphic EncryptionAlphabet · IEEE International Symposium on High-Performance Computer Architecture (HPCA)3 citationsJan 31AUM: Unleashing the Efficiency Potential of Shared Processors with Accelerator Units for LLM ServingAlibaba · IEEE International Symposium on High-Performance Computer Architecture (HPCA)2 citationsJan 31PinDrop: Breaking the Silence on SDCs in a Large-Scale FleetMeta · IEEE International Symposium on High-Performance Computer Architecture (HPCA)2 citationsJan 31Tawa: Automatic Warp Specialization for Modern GPUs with Asynchronous ReferencesNVIDIA · Proceedings of the CGO/Proceedings of the ... CGO2 citationsJan 31C³: CXL Coherence Controllers for Heterogeneous ArchitecturesNVIDIA · IEEE International Symposium on High-Performance Computer Architecture (HPCA)1 citationJan 31eGPU: Production-Scale Elastic Sharing Over 10,000 GPUsAlibaba · IEEE International Symposium on High-Performance Computer Architecture (HPCA)1 citationJan 31I-POP: Ignite Positive PrefetchersAlibaba · IEEE International Symposium on High-Performance Computer Architecture (HPCA)1 citationJan 31BitDecoding: Unlocking Tensor Cores for Long-Context LLMs with Low-Bit KV CacheMicrosoft · IEEE International Symposium on High-Performance Computer Architecture (HPCA)1 citationJan 31Proton: Towards Multi-level, Adaptive Profiling for TritonMeta, OpenAI · Proceedings of the CGO/Proceedings of the ... CGO1 citationJan 31Model Optimization for Multi-Camera 3D Detection and TrackingNVIDIA · arXiv (Cornell University)1 citationJan 31Cambricon-GS: An Accelerator for 3D Gaussian Splatting Training With Gaussian-Pixel Hybrid ParallelismCambricon Technologies · IEEE International Symposium on High-Performance Computer Architecture (HPCA)Jan 31QuCo: Efficient and Flexible Hardware-Driven Automatic Configuration of Tile Transfers in GPUsNVIDIA · IEEE International Symposium on High-Performance Computer Architecture (HPCA)Jan 31LEGO: Supporting LLM-Enhanced Games with One Gaming GPUMicrosoft · IEEE International Symposium on High-Performance Computer Architecture (HPCA)Jan 31GRANII: Selection and Ordering of Primitives in GRAph Neural Networks using Input InspectionNVIDIA · Proceedings of the CGO/Proceedings of the ... CGOJan 31Towards Threading the Needle of Debuggable Optimized BinariesAlphabet · Proceedings of the CGO/Proceedings of the ... CGOJan 31