Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Struct-Bench: A Benchmark for Differentially Private Structured Text GenerationGoogle DeepMind, Microsoft · neural information processing systems1 citationJan 1, 2025LiveCodeBench Pro: How Do Olympiad Medalists Judge LLMs in Competitive Programming?Google DeepMind · neural information processing systems1 citationJan 1, 2025Rendering-Aware Reinforcement Learning for Vector Graphics GenerationAlphabet · neural information processing systems1 citationJan 1, 2025Win Fast or Lose Slow: Balancing Speed and Accuracy in Latency-Sensitive Decisions of LLMsGoogle DeepMind, Microsoft, NVIDIA · neural information processing systems1 citationJan 1, 2025Hybrid Latent Reasoning via Reinforcement LearningAlphabet, Meta, SenseTime · neural information processing systems1 citationJan 1, 2025MLAN: Language-Based Instruction Tuning Preserves and Transfers Knowledge in Multimodal Language ModelsAlphabet · Workshop on Towards Knowledgeable Foundation Models (KnowFM)1 citationJan 1, 2025From English-Centric to Effective Bilingual: LLMs with Custom Tokenizers for Underrepresented LanguagesAlphabet · Ukrainian Natural Language Processing Workshop (UNLP)1 citationJan 1, 2025Optimizing Pre-Training Data Mixtures with Mixtures of Data Expert ModelsAlphabet · Proceedings of the conference - Association for Computational Linguistics. Meeting1 citationJan 1, 2025R3Mem: Bridging Memory Retention and Retrieval via Reversible CompressionAlphabet · Findings of the Association for Computational Linguistics: ACL (Findings of ACL)1 citationJan 1, 2025StableToolBench-MirrorAPI: Modeling Tool Environments as Mirrors of 7,000+ Real-World APIsAlphabet · Findings of the Association for Computational Linguistics: ACL (Findings of ACL)1 citationJan 1, 2025Subjective and Objective Quality Assessment of Banding Artifacts on Compressed VideosAlphabet · IEEE Transactions on Image Processing1 citationJan 1, 2025Design and Optimization of Efficient Digital Machine Learning Accelerators: An overview of architecture choices, efficient quantization, sparsity exploration, and system integration techniquesAlphabet, NVIDIA · IEEE Solid-State Circuits Magazine1 citationJan 1, 2025On Reference (In-)Determinacy in Natural Language InferenceAlphabet · Findings of the Association for Computational Linguistics: NAACL1 citationJan 1, 2025GeoCoder: Solving Geometry Problems by Generating Modular Code through Vision-Language ModelsGoogle DeepMind · Findings of the Association for Computational Linguistics: NAACL1 citationJan 1, 2025Prompting with Phonemes: Enhancing LLMs’ Multilinguality for Non-Latin Script LanguagesGoogle DeepMind · Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT)1 citationJan 1, 2025Minimal Evidence Group Identification for Claim VerificationAlphabet, Microsoft · Workshop on Trustworthy NLP (TrustNLP)1 citationJan 1, 2025FACT: Examining the Effectiveness of Iterative Context Rewriting for Multi-fact RetrievalAlphabet · Findings of the Association for Computational Linguistics: NAACL1 citationJan 1, 2025DreamWalk: Style Space Exploration Using Diffusion GuidanceAlphabet · Lecture notes in computer science1 citationJan 1, 2025Flow-Lenia: Emergent Evolutionary Dynamics in Mass Conservative Continuous Cellular AutomataGoogle DeepMind · Artificial Life1 citationJan 1, 2025Exploring the Accuracy and Privacy Tradeoff in AI-Driven Healthcare Through Differential PrivacyAlphabet · International Conference on Information Systems Security and Privacy (ICISSP)1 citationJan 1, 2025Cognate Production Using Character-Based Neural Machine Translation Without SegmentationAlphabet · IEEE Access1 citationJan 1, 2025On Vanishing Gradients, Over-Smoothing, and Over-Squashing in GNNs: Bridging Recurrent and Graph LearningGoogle DeepMind · neural information processing systems1 citationJan 1, 2025On the Coexistence and Ensembling of WatermarksGoogle DeepMind · neural information processing systems1 citationJan 1, 2025ConSim: Measuring Concept-Based Explanations’ Effectiveness with Automated SimulatabilityAlphabet · Proceedings of the conference - Association for Computational Linguistics. Meeting1 citationJan 1, 2025Improved Spectral Density Estimation via Explicit and Implicit DeflationAlphabet · Society for Industrial and Applied Mathematics eBooks1 citationJan 1, 2025