Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
TASTE2: Text-Aligned Speech Modeling and Deployment toward Full-Duplex Voice InteractionNVIDIA · arXiv (Cornell University)Sep 8Eliciting Weak-to-Strong Generalization with On-Policy Reverse DistillationMicrosoft · arXiv (Cornell University)Sep 8Agentic ML Exploration (A-MLE) for Ads RankingMeta · arXiv (Cornell University)Sep 8Curriculum Learning as Transport: Understanding Curricula with Wasserstein GeodesicsMicrosoft · arXiv (Cornell University)Sep 8Towards Standardized Evaluation of GPU Memory Safety with GMSBenchMicrosoft, NVIDIA · arXiv (Cornell University)Sep 8Charts Are Beyond Pixels: Probing for Layer-Wise Chart Understanding and EditingGoogle DeepMind · arXiv (Cornell University)Sep 8ExecCritic: Learn to Test, Test to Improve for Coding AgentsMicrosoft · arXiv (Cornell University)Sep 8It's Not RoPE that Creates Sinks: The Role of Self-Concentration and Value-Non-Mixing in AttentionPreferred Networks · arXiv (Cornell University)Sep 8Category-Native Solomonoff Approximation: From Algorithmic Geometry to Kernels, Operators, and InductionGoogle DeepMind · EntropySep 8Photonic Quantum-Enhanced Knowledge DistillationNVIDIA · Research SquareSep 8Multimodal alignment improves generalizability of genomic biomarker prediction in computational pathologyMicrosoft · Cell Reports MethodsSep 8Subagents vs Agent Skills: Executing Reusable Knowledge for Long-Horizon Agentic TasksMicrosoft · arXiv (Cornell University)Sep 7Elastic Horizon: Discovering the Effective Interaction Frontier in Agentic Reinforcement LearningAlibaba · arXiv (Cornell University)Sep 7FrogNano: Training a 4B Coding Agent via Online Task SynthesisMicrosoft · arXiv (Cornell University)Sep 7EigenLI: Spectral Approximations to Late InteractionMicrosoft · arXiv (Cornell University)Sep 7Protocol effects on feature-based hardware-Trojan detection across Trust-Hub familiesAlphabet · arXiv (Cornell University)Sep 7EAGER: Enrich-and-Align Generative Query Recommendation from Clicked Items in E-commerce SearchAlibaba · arXiv (Cornell University)Sep 7AgentIdeaBench: Benchmarking Scientific Ideation in the Agent EraNVIDIA · arXiv (Cornell University)Sep 7Large Discrete Policy: Advancing Explicit Behavior Modeling with Stochastic Iterative ScoringNVIDIA · arXiv (Cornell University)Sep 7P4Perf: Evaluating P4 Performance on dRMT NICs with DOCA Pipeline LanguageNVIDIA · Conference on Local Computer NetworksSep 7Causal evidence that language models use confidence to drive behaviourGoogle DeepMind · Nature Machine IntelligenceSep 7RoLA: Rotary-Positioned Low-Rank Linear Attention for Efficient Diffusion TransformersAlibaba · arXiv (Cornell University)Sep 6Inducing Emergent Misalignment from Reward Hacks with Iterative DPOGoogle DeepMind · arXiv (Cornell University)Sep 6Thinking with Cameras: Active Visual Reasoning via Dynamic Viewpoint Control for Surveillance Video UnderstandingSenseTime · arXiv (Cornell University)Sep 6Tight Bounds on the Cost of Adaptivity for the Meyerson SketchAlphabet · arXiv (Cornell University)Sep 6