Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Creating benchmarkable components to measure the quality ofAI-enhanced developer toolsAlphabet · arXiv (Cornell University)Apr 16, 2025Static to Dynamic Correlation ClusteringAlphabet · arXiv (Cornell University)Apr 16, 2025Hadamard Product in Deep Learning: Introduction, Advances and ChallengesGoogle DeepMind · IEEE Transactions on Pattern Analysis and Machine Intelligence35 citationsApr 15, 2025Bipartite Ranking From Multiple Labels: On Loss Versus Label AggregationAlphabet · arXiv (Cornell University)Apr 15, 2025MIST: A Co-Design Framework for Heterogeneous, Multi-Stage LLM InferenceGoogle DeepMind · arXiv (Cornell University)1 citationApr 14, 2025Learning to Prompt with Text Only Supervision for Vision-Language ModelsAlphabet · Proceedings of the AAAI Conference on Artificial Intelligence30 citationsApr 11, 2025MimiQ: Low-Bit Data-Free Quantization of Vision Transformers with Encouraging Inter-Head Attention SimilarityAlphabet · Proceedings of the AAAI Conference on Artificial Intelligence5 citationsApr 11, 2025QJL: 1-Bit Quantized JL Transform for KV Cache Quantization with Zero OverheadAlphabet · Proceedings of the AAAI Conference on Artificial Intelligence4 citationsApr 11, 2025High-Resolution Frame Interpolation with Patch-based Cascaded DiffusionAlphabet · Proceedings of the AAAI Conference on Artificial Intelligence3 citationsApr 11, 2025LLM Attributor: Interactive Visual Attribution for LLM GenerationAlphabet · Proceedings of the AAAI Conference on Artificial Intelligence3 citationsApr 11, 2025RLPF: Reinforcement Learning from Prediction Feedback for User Summarization with LLMsGoogle DeepMind · Proceedings of the AAAI Conference on Artificial Intelligence2 citationsApr 11, 2025Robust Multi-Objective Preference Alignment with Online DPOAlphabet, Google DeepMind · Proceedings of the AAAI Conference on Artificial Intelligence2 citationsApr 11, 2025Differentially Private Prototypes for Imbalanced Transfer LearningGoogle DeepMind · Proceedings of the AAAI Conference on Artificial Intelligence1 citationApr 11, 2025General Uncertainty Estimation with Delta VariancesGoogle DeepMind · Proceedings of the AAAI Conference on Artificial Intelligence1 citationApr 11, 2025How Do Position Encodings Affect Length Generalization? Case Studies On In-Context Function LearningAlphabet · Proceedings of the AAAI Conference on Artificial IntelligenceApr 11, 2025Bridging Algorithmic Information Theory and Machine Learning: Clustering, density estimation, Kolmogorov complexity-based kernels, and kernel learning in unsupervised learningGoogle DeepMind · Physica D Nonlinear Phenomena6 citationsApr 10, 2025Inexact Unlearning Needs More Careful Evaluations to Avoid a False Sense of PrivacyGoogle DeepMind · IEEE Conference on Secure and Trustworthy Machine Learning (SaTML)13 citationsApr 9, 2025SPLAT: A Framework for Optimised GPU Code-Generation for SParse reguLar ATtentionAlphabet, Microsoft · Proceedings of the ACM on Programming Languages3 citationsApr 9, 2025DART: A Principled Approach to Adversarially Robust Unsupervised Domain AdaptationAlphabet · IEEE Conference on Secure and Trustworthy Machine Learning (SaTML)2 citationsApr 9, 2025Locking Machine Learning Models into HardwareAlphabet, Google DeepMind · IEEE Conference on Secure and Trustworthy Machine Learning (SaTML)2 citationsApr 9, 2025Learning with User-Level Differential Privacy Under Fixed Compute BudgetsAlphabet · IEEE Conference on Secure and Trustworthy Machine Learning (SaTML)1 citationApr 9, 2025Avoiding Pitfalls for Privacy Accounting of Subsampled Mechanisms Under CompositionGoogle DeepMind · IEEE Conference on Secure and Trustworthy Machine Learning (SaTML)1 citationApr 9, 2025TxGemma: Efficient and Agentic LLMs for TherapeuticsAlphabet · arXiv (Cornell University)8 citationsApr 8, 2025Honest-Majority Threshold ECDSA with Batch Generation of Key-Independent PresignaturesAlphabet · IACR Communications in Cryptology4 citationsApr 8, 2025FEABench: Evaluating Language Models on Multiphysics Reasoning AbilityAlphabet · arXiv (Cornell University)Apr 8, 2025