Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
OfficeTown: Generative Organizations for Evolving Agent IntelligenceMicrosoft · arXiv (Cornell University)Oct 16Trojan Horse: Automated Repository Poisoning against Coding AgentsHugging Face · arXiv (Cornell University)Oct 16Sep-Latent: Separator Tokens as Natural Compression Anchors for Implicit ReasoningAlibaba · arXiv (Cornell University)Oct 16Humanity's Last Choice: A Common Evaluation Standard for Generative Engine OptimizationAlphabet, Google DeepMind · arXiv (Cornell University)Oct 16GR-OPD: Group-Relative Sequence Credit for On-Policy DistillationAlibaba · arXiv (Cornell University)Oct 16Self-Evolving Coding Agents via Harness-Policy OptimizationMicrosoft · arXiv (Cornell University)Oct 16Scalable Levenberg-Marquardt Training Framework for Deep LearningNVIDIA · arXiv (Cornell University)Oct 16JAXtari: High-Throughput and Easy-to-Modify Arcade Learning EnvironmentAlphabet · arXiv (Cornell University)Oct 16Reasoning Models Can Control Their Chains of Thought When They Tell Themselves ToAllen Institute for AI · arXiv (Cornell University)Oct 16ToPA: Block-wise Toeplitz Adaptation for Expressive and Efficient Fine-TuningAlibaba · arXiv (Cornell University)Oct 16Benchmarking and Enhancing Agentic Causal DiscoveryMicrosoft · arXiv (Cornell University)Oct 16TUPLE-Bench: Evaluating and Mitigating Privacy-Utility Trade-offs in Typed Decision ModelsHugging Face · arXiv (Cornell University)Oct 16ShadowFPT: Backdooring Federated Prompt Tuning via Shadow TriggersAlibaba · arXiv (Cornell University)Oct 16Sample, Mix, or Verify? Allocating Test-Time Compute under Imperfect VerifiersMicrosoft · arXiv (Cornell University)Oct 16Out of Support, Not Out of Reach: In-Context Learning with Unsupported Instructions and Composed TasksMicrosoft · arXiv (Cornell University)Oct 16"To Boldly Go Where No One Has Gone Before" What Has AI Let Developers Unlock?Microsoft · arXiv (Cornell University)Oct 16Understanding KV Residency in Production-Scale Agentic LLM WorkloadsMicrosoft · arXiv (Cornell University)Oct 16Nearly Linear Private Learning of Littlestone ClassesAlphabet · arXiv (Cornell University)Oct 16Predictable Scaling: Efficient Model Exploration via Scaling-Law PredictionMeta · arXiv (Cornell University)Oct 16Towards Unified Approaches in Self-Supervised Event Stream Modeling: Progress and ProspectsMicrosoft · Journal of Artificial Intelligence ResearchOct 16PW-Pruning: Phase-Aware Expert Selection for Speech-Centric Omni-Modal ModelsAlibaba · arXiv (Cornell University)Oct 15The “Humanization” Paradox of AI Shopping Assistants: The Double-Edged Sword Effect of Social Presence PerceptionMininglamp Technology · Journal of the Association for Information SystemsOct 15TokenEconomizer: Experience-Driven Black-Box Cost Control for Coding AgentsMicrosoft · arXiv (Cornell University)Oct 15At the Limits of Computability: Steps Towards Autonomous AI Discovery on the Busy Beaver FrontierGoogle DeepMind · arXiv (Cornell University)Oct 15Waypoint-OPD: Transferring RL-Induced Policy Shifts via Gated On-Policy DistillationAlibaba · arXiv (Cornell University)Oct 15