Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Read the Room, Read the Image: Understanding Indirect Speech Acts in Multimodal Visual ContextsLG AI Research · Findings of the Association for Computational Linguistics: ACL (Findings of ACL)Jan 1Lost in Stories: Consistency Bugs in Long Story Generation by LLMsMicrosoft · Findings of the Association for Computational Linguistics: ACL (Findings of ACL)Jan 1ToolRM: Towards Agentic Tool-Use Reward ModelingAlibaba · Findings of the Association for Computational Linguistics: ACL (Findings of ACL)Jan 1A Data-Efficient Path to Multilingual LLMs: Language Expansion via Post-training PARAM𝛥 Integration into Upcycled MoEAlibaba · Proceedings of the conference - Association for Computational Linguistics. MeetingJan 1Gradient-Guided Multi-Judge Prompt OptimizationMicrosoft · Proceedings of the conference - Association for Computational Linguistics. MeetingJan 1Beyond Rejection Sampling: Trajectory Fusion for Scaling Mathematical ReasoningMicrosoft · Findings of the Association for Computational Linguistics: ACL (Findings of ACL)Jan 1ShredBench: Evaluating the Semantic Reasoning Capabilities of Multimodal LLMs in Document ReconstructionAlibaba · Findings of the Association for Computational Linguistics: ACL (Findings of ACL)Jan 1AutoPKG: An Automated Framework for Dynamic E-commerce Product-Attribute Knowledge Graph ConstructionAlibaba · Findings of the Association for Computational Linguistics: ACL (Findings of ACL)Jan 1One Forward Beats Two: InnerZoom for Accurate and Efficient GUI GroundingAlibaba · Conference on Empirical Methods in Natural Language Processing (EMNLP)Jan 1LatentRevise: Learning from Zero-Hit ReasoningMicrosoft · Conference on Empirical Methods in Natural Language Processing (EMNLP)Jan 1DYNAMO: Dynamic Skill-Tool Evolution for Vision-Language AgentsAlibaba · Conference on Empirical Methods in Natural Language Processing (EMNLP)Jan 1HorizonRelight: Relighting Long-Horizon Videos Consistently via Diffusion TransformersNVIDIA · Lecture notes in computer scienceJan 1UniGP: Taming Diffusion Transformer for Prior-Preserved Unified Generation and PerceptionAlibaba · Lecture notes in computer scienceJan 1SEATauBench: Progressively Adapting Tool-Agent-User Evaluation Into Low-Resource Southeast Asian LanguagesCohere · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP)Jan 1A Physics-Grounded Benchmark for Multi-agent Dynamics in World ModelsNVIDIA · Lecture notes in computer scienceJan 1A 3GPP Perspective on Spectrum Sharing for 5G-to-6G Migration: From LTE-NR DSS Lessons to MRSS DesignNVIDIA · IEEE Communications Standards MagazineJan 1Security-Aware Spatiotemporal Video Enhancement for Resource-Limited Low-Altitude MonitoringAlibaba · IEEE Transactions on Consumer ElectronicsJan 1Revisiting Multi-Level Network Modeling and Simulation With Domain-Adapted LLMsNVIDIA · IEEE Transactions on Network Science and EngineeringJan 1MLVC: A Multi-platform Learned Video Codec for Real-World DeploymentMicrosoft · Lecture notes in computer scienceJan 1Focusing on What Matters: Saliency-Harnessing Accurate Routing for Diffusion MoEAlibaba · Lecture notes in computer scienceJan 1Jamming Detection at the Application Layer Using Deep Learning: A Transfer Learning ApproachTechnology Innovation Institute · IEEE Open Journal of the Communications SocietyJan 1MIMFlow: Integrating Masked Image Modeling with Normalizing Flows for End-to-End Image GenerationAlibaba · Lecture notes in computer scienceJan 1TryOnCrafter: Unleashing Camera Trajectories for Realistic Video Virtual Try-on via a Renderable 4D Try-on ProxyAlibaba · Lecture notes in computer scienceJan 1EchoStyle: Unlocking High-Fidelity Video Stylization with Reverse Data SynthesisAlibaba · Lecture notes in computer scienceJan 1Memory Makes the Difference: Evaluating How Different Memory Roles Shape Conversational AgentsMicrosoft · arXivJan 1