Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Skill-Decoupled On-Policy Expert Distillation for Unified Vision-Language ModelsAlibaba · arXiv (Cornell University)Oct 15ReFlow: Replay-Driven Decoupling for Efficient RL Post-TrainingAlibaba · arXiv (Cornell University)Oct 15Does On-Policy Distillation Need to Complete Every Rollout?Alibaba · arXiv (Cornell University)Oct 15MetaExp: Learning to Self-Improve through Experience via Verifiable Cross-Task Environment ScalingAlibaba · arXiv (Cornell University)Oct 15P2TPO: Structure-Aligned Page-to-Token Credit Assignment for Full-Page Document ParsingAlibaba · arXiv (Cornell University)Oct 15Learning to Steer LLM Agents and Multi-Agent SystemsAlibaba · arXiv (Cornell University)Oct 15Dendritic computation bridges neuroscience and foundation modelsAlibaba, MiniMax · Nature CommunicationsOct 15StreamErase: Towards Ultra-Fast Streaming Video ErasingAlibaba · arXiv (Cornell University)Oct 14What Makes a Social Agent's Interactions Effective? Learning from Verifiable Outcomes in Paired WorldsAlibaba · arXiv (Cornell University)Oct 14What Matters for World Modeling in Vision-Language-Action Models?Alibaba · arXiv (Cornell University)Oct 14Memory-Efficient Subspace Optimization with Residual Correction: From Momentum SGD to MuonAlibaba · arXiv (Cornell University)Oct 14Not All Criteria Matter: Importance-Aware Criteria Weighting for Rubric-Based LLM EvaluationAlibaba · arXiv (Cornell University)Oct 14ProtoArena: Can AI Agents Discover and Exploit Real-World Cross-Component Protocol Vulnerabilities?Alibaba · arXiv (Cornell University)Oct 13ScenePercept: Scenario-Conditioned Perceptual Assessment of Real-World User-Generated Content ImagesAlibaba · arXiv (Cornell University)Oct 13Socratic Skill Learning: Interleaving Teaching and Practice for Transferable Agent SkillsAlibaba · arXiv (Cornell University)Oct 13Multimodal Contrastive Representation Learning Based Voice Conversion for Electrolaryngeal Speech EnhancementAlibaba · arXiv (Cornell University)Oct 13RELAY: Selective Self-Distillation with Adaptive Hand-off for Rubric-Based Reinforcement LearningAlibaba · arXiv (Cornell University)Oct 13Heading Hierarchy Reconstruction for Complex and Long Documents with Level-Wise Multimodal MemoryAlibaba · arXiv (Cornell University)Oct 13Cowork-Gym: Synthesizing Interactive Work Environments at ScaleAlibaba · arXiv (Cornell University)Oct 13LoOPSD: On-Policy Self-Distillation with Looped TeachersAlibaba · arXiv (Cornell University)Oct 13LongForcing: Scaling Temporal Manifold Matching for Long-Horizon World RolloutAlibaba · arXiv (Cornell University)Oct 13X-Ragent: Tool-Augmented Reinforcement Learning for Evidence-Grounded Chest X-Ray ReasoningAlibaba · arXiv (Cornell University)Oct 12REVEAL: Reverse Engineering Agentic Latents from Frontier-Model TracesAlibaba · arXiv (Cornell University)Oct 12HyWe: Unifying Search and Operation for Real-World Hybrid Web TasksAlibaba · arXiv (Cornell University)Oct 12UniSGR-V2: Stepwise Target-Side Feature Anticipation and Crossing for Generative RecommendationAlibaba · arXiv (Cornell University)Oct 12