Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
SCOPE: Subspace Clustering with Online Per-Head Top-K Estimation for Sparse Video AttentionAlibaba · arXiv (Cornell University)Aug 13ATOBench: Tracing How Autonomous Penetration-Testing Agents Verify Vulnerabilities When Target Evidence LiesAlibaba · arXiv (Cornell University)Aug 13Structured Local Differential Modeling for AI-Generated Image DetectionAlibaba · arXiv (Cornell University)Aug 13HounsWorld: A Multimodal World Model for Hidden Patient-State Readout, Reconstruction, and SimulationAlibaba · arXiv (Cornell University)Aug 13Potential Applications of HBF in LLM Serving SystemsAlibaba · arXiv (Cornell University)Aug 13SelfVTON: Enhancing Virtual Try-On with Self-Supervised Cloth Detailing and Body AlignmentAlibaba · Proceedings - International Conference on Image ProcessingAug 13R-OVAR: Robust Open-Vocabulary Action Recognition for PracticalityAlibaba · Proceedings - International Conference on Image ProcessingAug 13Real-Time Image Restoration via Adaptive Decoupled Knowledge DistillationAlibaba · Proceedings - International Conference on Image ProcessingAug 13Squeeze Out Tokens from Sample for Finer-Grained Data GovernanceAlibaba · Proceedings - International Conference on Image ProcessingAug 13LiveAnimate: Stable Long-Form Streaming Human Animation in Real-TimeAlibaba · arXiv (Cornell University)Aug 12AVA-Encoder: Towards Agent-Native Video Representation LearningAlibaba · arXiv (Cornell University)Aug 12Ripple-Pivot Search: Active Parallel Decoding for Diffusion Large Language ModelsAlibaba · arXiv (Cornell University)Aug 12Measure, Don't Optimize: Forecasting Recovery in LLM UnlearningAlibaba · arXiv (Cornell University)Aug 11Whisper-Aware LLM: Self-Supervised Uncertainty Learning for Robust Whispered Speech RecognitionAlibaba · arXiv (Cornell University)Aug 11Dual-Loop Self-Evolution via Verifiable Emotion Feedback for Multi-Turn Empathetic DialogueAlibaba · arXiv (Cornell University)Aug 11SkillZip: Evaluation-Free Skill Compression for Self-Evolving Agents by Discovering Reusable StructureAlibaba · arXiv (Cornell University)Aug 11Can Released LLM Vocabularies Support Token-Level Estimation of Hidden Corpora?Alibaba · arXiv (Cornell University)Aug 11FADE: From Passive Verification to Active Discovery in Counterfactual Video UnderstandingAlibaba · arXiv (Cornell University)Aug 11REDAgentBench: Executable Red Teaming and Faithful Measurement of LLM Agent SystemsAlibaba · arXiv (Cornell University)Aug 11Balancing and Beyond: Communication-Centric Optimizations in Expert ParallelismAlibaba · ACM SIGCOMM Conference (SIGCOMM)Aug 11Rethinking Transparent TCP Replacement: Practical Lessons from SMC-R in the CloudAlibaba · ACM SIGCOMM Conference (SIGCOMM)Aug 11Draco-SLB: Supporting High-performance RDMA under Server Load Balancers for LLMAlibaba · ACM SIGCOMM Conference (SIGCOMM)Aug 11Networked Agent Memory and Causality Representation: Experiences towards Interpretable Cloud-Scale Root-CausingAlibaba · ACM SIGCOMM Conference (SIGCOMM)Aug 11PReCCL: Performant and Resilient Collective Communication via Integrated Inband Telemetry and Workload ReallocationAlibaba · ACM SIGCOMM Conference (SIGCOMM)Aug 11Anytest: Localizing the Root Cause of Hardware Transport Performance AnomaliesAlibaba · ACM SIGCOMM Conference (SIGCOMM)Aug 11