Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Understanding the Behaviors of Environment-aware Information RetrievalAlibaba · arXiv (Cornell University)Jun 15GD$^2$PO: Mitigating Multi-Reward Conflicts via Group-Dynamic reward-Decoupled Policy OptimizationAlibaba · arXiv (Cornell University)Jun 15SwiftCache: Efficient LLM Serving for Multi-turn Conversations with Heterogeneous KV Cache SharingAlibaba · arXiv (Cornell University)Jun 15FreeSonic: Training-Free Temporal-Aware Decoupled Attention for Precise Audio EditingAlibaba · arXiv (Cornell University)Jun 13Edu-Theater: A Data-Efficient Agent Framework for Scalable Learner Behavior Simulation through Staging Roll-CallAlibaba · arXiv (Cornell University)Jun 13SelfEvoSkill: Paired Execution Audits for Skill Revision without External Outcome SupervisionAlibaba · arXiv (Cornell University)Jun 12ClinHallu: A Benchmark for Diagnosing Stage-Wise Hallucinations in Medical MLLM ReasoningAlibaba · arXiv (Cornell University)Jun 12Compressing Image Style Training into a Single Model ForwardAlibaba · arXiv (Cornell University)Jun 11Dense Supervision, Sparse Updates: On the Sparsity and Geometry of On-Policy DistillationAlibaba · arXiv (Cornell University)Jun 11SkillChain: Closing the Loop on Skill Evolution for Image-Based E-Commerce AI AssistantsAlibaba · arXiv (Cornell University)Jun 11ReSum: Synergizing LLM Reasoning and Summarization with Reinforcement LearningAlibaba · arXiv (Cornell University)Jun 11APPO: Agentic Procedural Policy OptimizationAlibaba · arXiv (Cornell University)Jun 10Verifiable Environments Are LEGO Bricks: Recursive Composition for Reasoning GeneralizationAlibaba · arXiv (Cornell University)Jun 10RLCSD: Reinforcement Learning with Contrastive On-Policy Self-DistillationAlibaba · arXiv (Cornell University)Jun 10Breaking Entropy Bounds: Accelerating RL Training via MTP with Rejection SamplingAlibaba · arXiv (Cornell University)Jun 10How Does Reasoning Flow? Tracing Attention-Induced Information Flow for Targeted RL in LLMsAlibaba · arXiv (Cornell University)Jun 9From Context-Aware to Conflict-Aware: Generalizing Contrastive Decoding for Knowledge Conflict in LLMsAlibaba · arXiv (Cornell University)Jun 9Role-Agent: Bootstrapping LLM Agents via Dual-Role EvolutionAlibaba · arXiv (Cornell University)Jun 9K-Forcing: Joint Next-K-Token Decoding via Push-Forward Language ModelingAlibaba · arXiv (Cornell University)Jun 9AniFeats: Animate 3D Feature Meshes for Character Video GenerationAlibaba · IEEE Transactions on Visualization and Computer GraphicsJun 9From Slice to Sequence: Autoregressive Tracking Transformer for Consistent 3-D Lymph Node Detection in CT ScansAlibaba · IEEE Transactions on Medical ImagingJun 9BareWave: Waveform-Native Flow-Matching Text-to-SpeechAlibaba · arXiv (Cornell University)Jun 8HybridSkipList+: Rethinking Distributed Skiplist With Hybrid RDMA and CachingAlibaba · IEEE Transactions on ComputersJun 8RASO: Role-Aware Shared Reflection for Multi-Agent Orchestration in E-Commerce Long-Horizon PlanningAlibaba · Proceedings of the International Conference on Automated Planning and SchedulingJun 8TQR : Modelling user temporal preference for effective question routingAlibaba · Neural NetworksJun 8