Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
From Distributions to Stochastic Processes: Neural Approximation of Measure-Valued MapsAlibaba · arXiv (Cornell University)Sep 27TeacherGRPO: Closing the Capacity Gap in Reasoning Distillation via Teacher AlignmentMicrosoft · arXiv (Cornell University)Sep 27Finite Probes Suffice: Identifiability and Universality for Weight-Space LearningNVIDIA · arXiv (Cornell University)Sep 27Learning Dynamics of Continual Learning: A Unified View of Data Attribution, Forgetting, and Plasticity LossGoogle DeepMind · arXiv (Cornell University)Sep 27On Device Agentic Operation Caches -- Classifier-Centric NL-to-Action GenerationMicrosoft · arXiv (Cornell University)Sep 27DuraS2ST: Chain-of-Thought and Reinforcement Learning for Duration-Aligned Speech-to-Speech TranslationMicrosoft · arXiv (Cornell University)Sep 27QwenGyre: An Elastic Reinforcement Learning Framework for Training xLong-Horizon AgentsAlibaba · arXiv (Cornell University)Sep 27Selecting Diverse SFT Traces Improves Post-RL GeneralizationAlphabet · arXiv (Cornell University)Sep 27Splicing Hints into Reasoning Traces: Learning from Demonstrator FailuresAlphabet · arXiv (Cornell University)Sep 26Predict Before You Modify: Learning Improvement Dynamics for Self-Evolving AgentsMicrosoft · arXiv (Cornell University)Sep 26Certainty Is Not Just Correctness: Rethinking Token-Level Certainty in LLM ReasoningAlibaba · arXiv (Cornell University)Sep 26Architectural Paradigm Shifts in Data Center Servers: A Review of Workload Offloading, Security Isolation, and the Rise of Infrastructure Processing Units (IPUs)Alphabet · Journal of Computer Science and Technology StudiesSep 26Neural Dynamics as the Composition of Quantized UnitsSakana AI · arXiv (Cornell University)Sep 26Efficient Dynamic Algorithms for Graph Neural Networks with Non-Linear PropagationAlphabet · arXiv (Cornell University)Sep 26How Far Do Persona Effects Generalize in Language Models?Microsoft · arXiv (Cornell University)Sep 26MM-OPD: Towards One More Bottleneck Between Perception and ReasoningAlibaba · arXiv (Cornell University)Sep 26ScopeIF: Improving Scope-Aware Precise Instruction-Following in Large Language Models via Graded Reward ModelingZhipu AI · arXiv (Cornell University)Sep 26From Anomalies to Failures: Constructing Causal Error Graphs for Agentic Trace DiagnosisZhipu AI · arXiv (Cornell University)Sep 26KeyRec: Bounded Visual Memory for Streaming and Long-Video UnderstandingAlphabet · arXiv (Cornell University)Sep 26Opening LLM Judges: Recovering Preference Signals Beyond the Final VerdictMicrosoft · arXiv (Cornell University)Sep 26DraftAttention2: Fast Video Diffusion with Low-Resolution-Guided Mixed-Precision AttentionNVIDIA · arXiv (Cornell University)Sep 26FA-Bench: A Benchmark for Phone- and Word-Level Timestamp Accuracy in Forced Alignment and ASR on Clean and Noisy SpeechMicrosoft · arXiv (Cornell University)Sep 26Permutation-Equivariant Flow Matching for Alignment-Free Neural Weight GenerationNVIDIA · arXiv (Cornell University)Sep 26Allspark: Weak to Strong Transfer via Alternating Chain of ThoughtMicrosoft · arXiv (Cornell University)Sep 26Representation Editing for Multimodal Test-Time AdaptationAlibaba · arXiv (Cornell University)Sep 26