Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Beyond Entropy: Self-Diagnostic Multi-Role Token Optimization for Video ReasoningAlibaba · arXiv (Cornell University)Oct 2MoLE: Mixture of Latent Experts for Complementary Visual ReasoningAlibaba · arXiv (Cornell University)Oct 1DeFA: Dependency-Guided Failure Attribution for LLM AgentsAlibaba · arXiv (Cornell University)Oct 1How the Audit Rule Shapes Faithful Factor Explanations in LLMsAlibaba · arXiv (Cornell University)Oct 1OverAct: Measuring and Mitigating Proactive Over-Authorization in LLM Tool-Calling AgentsAlibaba · arXiv (Cornell University)Oct 1My FAULT: Self-Diagnosis as Credit Assignment in Self-Evolving Agentic Reinforcement LearningAlibaba · arXiv (Cornell University)Oct 1MatrixReward: Reward from Rubric Matrix for Open-Ended GenerationAlibaba · arXiv (Cornell University)Sep 30Herschel: Continuous Optimization of Production LLM Inference through On-Demand ProfilingAlibaba · arXiv (Cornell University)Sep 30SCIC: Scope- and Codebook-Aware Instruction Conditioning for Speaker-Adapted Expressive TTSAlibaba · arXiv (Cornell University)Sep 30Capture the lifecycle: KV Cache management in ReAct Agents with KVTetherAlibaba · arXiv (Cornell University)Sep 30ID Balancing: Stable Training of Extremely Sparse MoE via PID-Based Load ControlAlibaba · arXiv (Cornell University)Sep 30StateTree: Enhancing Long-Term Dialogue Reasoning via Reinforcement LearningAlibaba · arXiv (Cornell University)Sep 30AdaM-Rec: Adaptive Modality Routing for Multimodal RecommendationAlibaba · arXiv (Cornell University)Sep 29When Upstream Messages Override Correct Answers: A Controlled Study of Multi-Agent LLM CollaborationAlibaba · arXiv (Cornell University)Sep 29How Can Recommendation Feedback Evolve Agent Memory?Alibaba · arXiv (Cornell University)Sep 29Fine-Tuning on Self-Generated and Reward-Weighted Data: Learning Dynamics, Convergence Rates, and Benefits of Off-PolicynessAlibaba · arXiv (Cornell University)Sep 29A Few GPUs, A Whole Lotta Scale: Faithful LLM Training Emulation with CrystalLLMAlibaba · ACM SIGOPS Symposium on Operating Systems Principles (SOSP)Sep 28Beyond Verbalized Confidence: Calibrating Reasoners with Differentiable ReadoutsAlibaba · arXiv (Cornell University)Sep 28Teacher-Student Gaps Are Not Enough: Outcome-Guided On-Policy Distillation for Multi-Turn Autonomous AgentsAlibaba · arXiv (Cornell University)Sep 28PEARL: Adaptive Prefill-Decode Execution with Elasticity for Agentic Reinforcement LearningAlibaba · arXiv (Cornell University)Sep 28Reward-Aligned Reweighting for On-Policy DistillationAlibaba · arXiv (Cornell University)Sep 28Token-Disentangled Latent Test-Time Scaling for Vision-Language ReasoningAlibaba · arXiv (Cornell University)Sep 28Muon Sublates the Edge of Stability in LLM PretrainingAlibaba · arXiv (Cornell University)Sep 28TSGate: Timestep-Aware Gated Attention for Diffusion TransformersAlibaba · arXiv (Cornell University)Sep 28When Confidence Rises Too Early: Detecting Shortcut Reasoning via Premature Answer CommitmentAlibaba · arXiv (Cornell University)Sep 28