Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
TAC: Timestamped Audio CaptioningOpenAI · arXiv (Cornell University)Feb 17Kunlun: Establishing Scaling Laws for Massive-Scale Recommendation Systems through Unified Architecture DesignMeta, OpenAI · arXiv (Cornell University)Feb 10Kunlun: Establishing Scaling Laws for Massive-Scale Recommendation Systems through Unified Architecture DesignMeta, OpenAI · arXiv (Cornell University)Feb 10Reducing Isotropy and Volume to KLS: Faster Rounding and Volume AlgorithmsOpenAI · Journal of the ACM2 citationsFeb 7Proton: Towards Multi-level, Adaptive Profiling for TritonMeta, OpenAI · IEEE/ACM International Symposium on Code Generation and Optimization (CGO)1 citationJan 31A benchmark of expert-level academic questions to assess AI capabilitiesAlphabet, Anthropic, Cohere, Google DeepMind, Microsoft, OpenAI · Nature61 citationsJan 28Thought-Transfer: Indirect Targeted Poisoning Attacks on Chain-of-Thought Reasoning ModelsAnthropic, Google DeepMind, OpenAI · arXiv (Cornell University)Jan 27GDPVAL: Evaluating AI Model Performance on Real-World Economically Valuable TasksOpenAI · SuperIntelligence - Robotics - Safety & Alignment6 citationsJan 21A Resource-Rational Account of Human Eye Movements During Immersive Visual SearchOpenAI · Open Mind1 citationJan 1Imaginative Perception Tokens Enhance Spatial Reasoning in Multimodal Language ModelsAllen Institute for AI, Microsoft, OpenAI · Lecture notes in computer scienceJan 1Linear Layouts: Robust Code Generation of Efficient Tensor Computation Using F_2OpenAI · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)3 citationsDec 11, 2025Improving Streaming ASR via Differentially Private Fusion of Data from Multiple SourcesAlphabet, Google DeepMind, OpenAI · IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)Dec 6, 2025TransAct V2: Lifelong User Action Sequence Modeling on Pinterest RecommendationOpenAI · ACM International Conference on Information and Knowledge Management (CIKM)4 citationsNov 8, 2025Trust but Verify: Programmatic VLM Evaluation in the WildOpenAI · IEEE/CVF International Conference on Computer Vision (ICCV)1 citationOct 19, 2025Adam or Gauss-Newton? A Comparative Study In Terms of Basis Alignment and SGD NoiseOpenAI · arXiv (Cornell University)Oct 15, 2025Mind the Abstraction Gap: Bringing Equality Saturation to Real-World ML CompilersOpenAI · Proceedings of the ACM on Programming Languages3 citationsOct 9, 2025Open ASR Leaderboard: Towards Reproducible and Transparent Multilingual and Long-Form Speech Recognition EvaluationHugging Face, Mistral AI, NVIDIA, OpenAI · arXiv (Cornell University)1 citationOct 8, 2025Expand Neurons, Not ParametersOpenAI · arXiv (Cornell University)Oct 6, 2025Mercury: Unlocking Multi-GPU Operator Optimization for LLMs via Remote Memory SchedulingMeta, OpenAI · ACM SIGOPS Symposium on Operating Systems Principles (SOSP)5 citationsOct 1, 2025Ranking by engagement and non‐engagement signals: Learnings from industryOpenAI · Annals of the New York Academy of Sciences2 citationsAug 20, 2025DV365: Extremely Long User History Modeling at InstagramMeta, OpenAI · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)3 citationsAug 3, 2025Deliberative Alignment: Reasoning Enables Safer Language ModelsOpenAI · SuperIntelligence - Robotics - Safety & Alignment35 citationsJul 20, 2025Improving Diffusion Inverse Problem Solving with Decoupled Noise AnnealingNVIDIA, OpenAI · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings26 citationsJun 10, 2025Vision-Language Models Do Not Understand NegationOpenAI · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings22 citationsJun 10, 2025Linear Layouts: Robust Code Generation of Efficient Tensor Computation Using $\mathbb{F}_2$OpenAI · arXiv (Cornell University)May 28, 2025