Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Parallelizing the Approximate Minimum Degree Ordering Algorithm: Strategies and EvaluationNVIDIA · Society for Industrial and Applied Mathematics eBooksFeb 13Quantum Eigenvalue ProcessingMicrosoft · SIAM Journal on Computing7 citationsFeb 12Adaptive Debiasing Tsallis Entropy for Test-Time AdaptationAlibaba · arXiv (Cornell University)5 citationsFeb 12On-Policy Context Distillation for Language ModelsMicrosoft · arXiv (Cornell University)3 citationsFeb 12Olmix: A Framework for Data Mixing Throughout LM DevelopmentAllen Institute for AI · arXiv (Cornell University)2 citationsFeb 12Choose Your Agent: Tradeoffs in Adopting AI Advisors, Coaches, and Delegates in Multi-Party NegotiationAlphabet · arXiv (Cornell University)1 citationFeb 12Keeping a Secret Requires a Good Memory: Space Lower-Bounds for Private AlgorithmsAlphabet · arXiv (Cornell University)Feb 12Temperature as a Meta-Policy: Adaptive Temperature in LLM Reinforcement LearningMicrosoft · arXiv (Cornell University)Feb 12When Should LLMs Be Less Specific? Selective Abstraction for Reliable Long-Form Text GenerationNVIDIA · arXiv (Cornell University)Feb 12Towards On-Policy SFT: Distribution Discriminant Theory and its Applications in LLM TrainingMicrosoft · arXiv (Cornell University)Feb 12SIGHT: Reinforcement Learning with Self-Evidence and Information-Gain Diverse Branching for Search AgentAlibaba · arXiv (Cornell University)Feb 12Enhancing MMDiT-Based Text-to-Image Models for Similar Subject GenerationMicrosoft · IEEE Transactions on Pattern Analysis and Machine Intelligence5 citationsFeb 11GeoGR: Enabling Spatio-Temporal Aware Industrial-scale Generative POI RecommendationsAlibaba · arXiv (Cornell University)1 citationFeb 11Data Repetition Beats Data Scaling in Long-CoT Supervised Fine-TuningMistral AI, NVIDIA · arXiv (Cornell University)1 citationFeb 11SoftMatcha 2: A Fast and Soft Pattern Matcher for Trillion-Scale CorporaSakana AI · arXiv (Cornell University)1 citationFeb 11ISD-Agent-Bench: A Comprehensive Benchmark for Evaluating LLM-based Instructional Design AgentsUpstage · arXiv (Cornell University)1 citationFeb 11VTC: DNN Compilation with Virtual Tensors for Data Movement EliminationMicrosoft · arXiv (Cornell University)Feb 11Next-token Preference: Scaling Unsupervised Reward ModelingMicrosoft · arXiv (Cornell University)Feb 11The Magic Correlations: Understanding Knowledge Transfer from Pretraining to Supervised Fine-TuningAlphabet · arXiv (Cornell University)Feb 11Are Aligned Large Language Models Still Misaligned?Microsoft · arXiv (Cornell University)Feb 11Stress Tests REVEAL Fragile Temporal and Visual Grounding in Video-Language ModelsAlphabet, NVIDIA · arXiv (Cornell University)Feb 11RISE: Self-Improving Robot Policy with Compositional World ModelHorizon Robotics · arXiv (Cornell University)Feb 11Chamfer-Linkage for Hierarchical Agglomerative ClusteringAlphabet · arXiv (Cornell University)Feb 11Kunlun: Establishing Scaling Laws for Massive-Scale Recommendation Systems through Unified Architecture DesignMeta, OpenAI · arXiv (Cornell University)3 citationsFeb 10AI-Assisted Scientific Assessment: A Case Study on Climate ChangeGoogle DeepMind · arXiv (Cornell University)3 citationsFeb 10