Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
"Movement is your ability to engage with the world": Opportunities for Reflecting, Revealing, and Hiding Visible and Invisible Disabilities Through Avatar Movement in VRMicrosoft · International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS)Oct 23A Lightweight Deep Learning Framework for Real-Time Cyberattack Detection in Internet of Things NetworksMicrosoft · Computer Science BulletinOct 23What the Specification Did Not Say: Specification-Oracle Gaps in Test-Graded Coding-Agent BenchmarksAlibaba · arXiv (Cornell University)Oct 22Sensorimotor Alignment: A Platonic Proxy for Embodied Policy SuccessNVIDIA · arXiv (Cornell University)Oct 22Scaling Laws for Safety ProbesAnthropic · arXiv (Cornell University)Oct 22Towards User Simulators That Stay On Task And Still Sound NaturalMicrosoft · arXiv (Cornell University)Oct 22Reverse-KL Policy Distillation: Statistical Guarantees and LimitsAlphabet · arXiv (Cornell University)Oct 22UserArena: Benchmarking Interactive User Simulation for Long-Horizon Shopping AgentsAlibaba · arXiv (Cornell University)Oct 22ResGVC: Residual Coding for High-fidelity Generative Video CompressionAlibaba · arXiv (Cornell University)Oct 22Building Presentation Intelligence for High-Quality Slide GenerationAlibaba · arXiv (Cornell University)Oct 22How Depth Growth and Looping Shape In-Context Learning in Language ModelsAlphabet · arXiv (Cornell University)Oct 22Everything everywhere all‐at‐once? AI , expertise–output decoupling, and new entrepreneurial strategy trade‐offs in the age of abundanceAlphabet, Microsoft · Strategic Entrepreneurship JournalOct 22Towards Faithful Data Selection: Statistical Limits and Scalable AlgorithmsGoogle DeepMind · arXiv (Cornell University)Oct 21PLE: Physical-Engineering's Last ExamAlphabet · arXiv (Cornell University)Oct 21Short-to-Long: Detached Features for Long-Horizon Value LearningNVIDIA · arXiv (Cornell University)Oct 21Bridging Foundational and Agentic Capabilities of LLMsMicrosoft · arXiv (Cornell University)Oct 21Characterizing Social Media Behavior under Limited Observability: Retweet Link Prediction and Multimodal Intention RecognitionAlphabet · arXiv (Cornell University)Oct 21SWE-DIVE: Multi-Turn Coding Agent Evaluation with Dynamic User SimulationMicrosoft · arXiv (Cornell University)Oct 21Right On Q: Recovering Reasoning under Quantization via Self-Generated TrajectoriesMeta · arXiv (Cornell University)Oct 21Do Coding Agents Understand Why Systems Crash---and Why They Don't?Alphabet, Google DeepMind · arXiv (Cornell University)Oct 21ProactEgoMem: Benchmarking Proactive Memory in Streaming Egocentric VideoAlphabet · arXiv (Cornell University)Oct 21Building Auditable AI Text Detectors with Sparse AutoencodersGoogle DeepMind · arXiv (Cornell University)Oct 21Derivation Depth Exposes Reasoning Failures in Tool-Augmented Music AnalysisAlphabet · arXiv (Cornell University)Oct 21Omnilingual-MsASR: Benchmarking Multimodal, Multilingual, Long-Form Multi-Speaker ASR of Omni Large Language ModelsAlibaba · arXiv (Cornell University)Oct 21CausalAlign: Causal Alignment of Video Generation Models for Consistent Ego-MotionNVIDIA · arXiv (Cornell University)Oct 21