Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
IllumiCraft: Unified Geometry and Illumination Diffusion for Controllable Video GenerationAlphabet, Google DeepMind · neural information processing systemsJan 1, 2025Gatekeeper: Improving Model Cascades Through Confidence TuningAlphabet, Google DeepMind · neural information processing systemsJan 1, 2025Meta-learning how to Share Credit among Macro-ActionsGoogle DeepMind · neural information processing systemsJan 1, 2025MV-CoLight: Efficient Object Compositing with Consistent Lighting and Shadow GenerationAlphabet, Google DeepMind · neural information processing systemsJan 1, 2025Beyond Verifiable Rewards: Scaling Reinforcement Learning in Language Models to Unverifiable DataGoogle DeepMind · neural information processing systemsJan 1, 2025Asymmetric REINFORCE for off-Policy Reinforcement Learning: Balancing positive and negative rewardsGoogle DeepMind, Meta · neural information processing systemsJan 1, 2025Scalable In-context Ranking with Generative ModelsAlphabet, Google DeepMind · neural information processing systemsJan 1, 2025SegMASt3R: Geometry Grounded Segment MatchingGoogle DeepMind · neural information processing systemsJan 1, 2025IF-Guide: Influence Function-Guided Detoxification of LLMsAnthropic, Google DeepMind · neural information processing systemsJan 1, 2025On Traceability in $\ell_p$ Stochastic Convex OptimizationGoogle DeepMind · neural information processing systemsJan 1, 2025Constructing an Optimal Behavior Basis for the Option KeyboardGoogle DeepMind · neural information processing systemsJan 1, 2025AmorLIP: Efficient Language-Image Pretraining via AmortizationAlphabet, Google DeepMind · neural information processing systemsJan 1, 2025SkillVerse : Assessing and Enhancing LLMs with Tree EvaluationGoogle DeepMind · Annual Meeting of the Association for Computational Linguistics (ACL)Jan 1, 2025Keep Guessing? When Considering Inference Scaling, Mind the BaselinesAlphabet, Google DeepMind · Findings of the Association for Computational Linguistics: NAACLJan 1, 2025Chasing Random: Instruction Selection Strategies Fail to GeneralizeAlphabet, Google DeepMind · Findings of the Association for Computational Linguistics: NAACLJan 1, 2025Tomato, Tomahto, Tomate: Do Multilingual Language Models Understand Based on Subword-Level Semantic Concepts?Google DeepMind · Findings of the Association for Computational Linguistics: NAACLJan 1, 2025Overcoming Sparsity Artifacts in Crosscoders to Interpret Chat-TuningGoogle DeepMind · neural information processing systemsJan 1, 2025Dimension-free Score Matching and Time Bootstrapping for Diffusion ModelsGoogle DeepMind · neural information processing systemsJan 1, 2025Analyzing Similarity Metrics for Data Selection for Language Model PretrainingAlphabet, Google DeepMind · neural information processing systemsJan 1, 20253D Interaction Geometric Pre-training for Molecular Relational LearningGoogle DeepMind · neural information processing systemsJan 1, 2025Informed Correctors for Discrete Diffusion ModelsGoogle DeepMind, Microsoft · neural information processing systemsJan 1, 2025