Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Accelerating 3D Gaussian Splatting using Tensor CoresMicrosoft · arXiv (Cornell University)May 18Echoes of Citations: Automated Extraction of Claims from Full Scientific PapersAllen Institute for AI, Google DeepMind, Microsoft · Proceedings of the AAAI Symposium SeriesMay 18CommitDistill: A Lightweight Knowledge-Centric Memory Layer for Software RepositoriesMicrosoft · arXiv (Cornell University)May 18EPIC: Abstraction and Polymorphism of In-Network Collectives on EthernetAlibaba · arXiv (Cornell University)May 18A Data-Efficient Path to Multilingual LLMs: Language Expansion via Post-training PARAM$Δ$ Integration into Upcycled MoEAlibaba · arXiv (Cornell University)May 18Interactive Evaluation Requires a Design ScienceMicrosoft · arXiv (Cornell University)May 18Incantation: Natural Language as the Action Interface for Multi-Entity Video World ModelsNVIDIA · arXiv (Cornell University)May 18See What I Mean: Aligning Vision and Language Representations for Video Fine-grained Object UnderstandingAlibaba · arXiv (Cornell University)May 18Continuous Diffusion Scales Competitively with Discrete Diffusion for LanguageNVIDIA · arXiv (Cornell University)May 18Kangaroo: Efficient Lossless Floating-Point Compression via Dynamic Reference SelectionAlibaba · Proceedings of the ACM on Management of DataMay 18DFLOP: A Data-driven Framework for Multimodal LLM Training Pipeline OptimizationMicrosoft · Proceedings of the ACM on Management of DataMay 18Sketch-based Secure Query Processing for Streaming DataAlibaba · Proceedings of the ACM on Management of DataMay 18Understanding the Impact of Data Noise in Federated Learning: [Experiments & Analysis]Microsoft · Proceedings of the ACM on Management of DataMay 18CoDec: Prefix-Shared Decoding Kernel for LLMsAlibaba · Proceedings of the ACM on Management of DataMay 18Meep Hashing: Ultrafast and Compact Minimal Perfect Hashing for Practical Large-Scale Lookup SystemsAlphabet · Proceedings of the ACM on Management of DataMay 18MoDora: Tree-Based Semi-Structured Document Analysis SystemMicrosoft · Proceedings of the ACM on Management of DataMay 18LSHAlign: All-Pair Near-Duplicate Text Alignment via Locality-Sensitive HashingAlibaba · Proceedings of the ACM on Management of DataMay 18100x Cost & Latency Reduction: Performance Analysis of AI Query Approximation using Lightweight Proxy Models: [Experiments & Analysis]Alphabet · Proceedings of the ACM on Management of DataMay 18AMATA: Adaptive Multi-Agent Trajectory Alignment for Knowledge-Intensive Question AnsweringAlibaba · arXiv (Cornell University)May 17Taming "Zombie'' Agents: A Markov State-Aware Framework for Resilient Multi-Agent EvolutionAlibaba · arXiv (Cornell University)May 17VeriCache: Turning Lossy KV Cache into Lossless LLM InferenceMicrosoft · arXiv (Cornell University)May 17GraphMind: From Operational Traces to Self-Evolving Workflow AutomationMicrosoft · arXiv (Cornell University)May 17DISA: Offline Importance Sampling for Distribution-Matching LLM-RLAlibaba · arXiv (Cornell University)May 17LLMForge: Multi-Backend Hardware-Aware Neural Architecture Search with Infinite-Head Attention for Edge Language ModelsAlphabet · arXiv (Cornell University)May 17Trust No Tool: Evaluating and Defending LLM Agents under Untrusted Tool FeedbackAlibaba · arXiv (Cornell University)May 17