Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
GENMAC: Compositional Text-to-Video Generation with Multi-Agent CollaborationMicrosoft · Proceedings of the AAAI Conference on Artificial Intelligence5 citationsMar 14Reasoning with Exploration: An Entropy PerspectiveMicrosoft · Proceedings of the AAAI Conference on Artificial Intelligence5 citationsMar 14ClearAIR: A Human-Visual-Perception-Inspired All-in-One Image RestorationHorizon Robotics · Proceedings of the AAAI Conference on Artificial Intelligence5 citationsMar 14MonoDream: Monocular Vision-Language Navigation with Panoramic DreamingHorizon Robotics · Proceedings of the AAAI Conference on Artificial Intelligence4 citationsMar 14LLM2CLIP: Powerful Language Model Unlocks Richer Cross-Modality RepresentationMicrosoft · Proceedings of the AAAI Conference on Artificial Intelligence3 citationsMar 14CompTrack: Information Bottleneck-Guided Low-Rank Dynamic Token Compression for Point Cloud TrackingMininglamp Technology · Proceedings of the AAAI Conference on Artificial Intelligence3 citationsMar 14Schema-Guided Scene-Graph Reasoning Based on Multi-Agent Large Language Model SystemMicrosoft, NVIDIA · Proceedings of the AAAI Conference on Artificial Intelligence3 citationsMar 14DocR1: Evidence Page-Guided GRPO for Multi-Page Document UnderstandingiFlytek · Proceedings of the AAAI Conference on Artificial Intelligence3 citationsMar 14Evaluating, Synthesizing, and Enhancing for Customer Support ConversationAlibaba · Proceedings of the AAAI Conference on Artificial Intelligence2 citationsMar 14DCRNet: Delayed Conversion Modeling Based Personalized Flight Itinerary Ranking NetworkAlibaba · Proceedings of the AAAI Conference on Artificial Intelligence2 citationsMar 14LLM Collaborative Filtering: User-Item Graph as New LanguageAlibaba · Proceedings of the AAAI Conference on Artificial Intelligence2 citationsMar 14SpeakerLM: End-to-End Versatile Speaker Diarization and Recognition with Multimodal Large Language ModelsAlibaba · Proceedings of the AAAI Conference on Artificial Intelligence2 citationsMar 14Marginalized Generalized IoU (MGIoU): A Unified Objective Function for Optimizing Convex Parametric ShapesNVIDIA · Proceedings of the AAAI Conference on Artificial Intelligence2 citationsMar 14Where and What Matters: Sensitivity-Aware Task Vectors for Many-Shot Multimodal In-Context LearningAlibaba · Proceedings of the AAAI Conference on Artificial Intelligence2 citationsMar 14Efficient and Adaptive Simultaneous Speech Translation with Fully Unidirectional ArchitectureAlibaba · Proceedings of the AAAI Conference on Artificial Intelligence2 citationsMar 14Video SimpleQA: Towards Factuality Evaluation in Large Video Language ModelsAlibaba · Proceedings of the AAAI Conference on Artificial Intelligence2 citationsMar 14Margin-Aware Preference Optimization for Aligning Diffusion Models Without ReferenceHugging Face · Proceedings of the AAAI Conference on Artificial Intelligence2 citationsMar 14HiTVideo: Hierarchical Tokenizers for Enhancing Text-to-Video Generation with Autoregressive Large Language ModelsMicrosoft · Proceedings of the AAAI Conference on Artificial Intelligence2 citationsMar 14Q Cache: Visual Attention Is Valuable in Less than Half of Decode Layers for Multimodal Large Language ModelAlibaba · Proceedings of the AAAI Conference on Artificial Intelligence2 citationsMar 14MMG-Vid: Maximizing Marginal Gains at Segment-level and Token-level for Efficient Video LLMsAlibaba · Proceedings of the AAAI Conference on Artificial Intelligence2 citationsMar 14A Course Correction in Steerability Evaluation: Revealing Miscalibration and Side Effects in LLMsMicrosoft · Proceedings of the AAAI Conference on Artificial Intelligence1 citationMar 14Breaking Model Lock-in: Cost-Efficient Zero-Shot LLM Routing via a Universal Latent SpaceiFlytek · Proceedings of the AAAI Conference on Artificial Intelligence1 citationMar 14Enhancing Spatial Reasoning Through Visual and Textual ThinkingAlibaba · Proceedings of the AAAI Conference on Artificial Intelligence1 citationMar 14AdaCuRL: Adaptive Curriculum Reinforcement Learning with Invalid Sample Mitigation and Historical RevisitingAlibaba · Proceedings of the AAAI Conference on Artificial Intelligence1 citationMar 14Omni-Effects: Unified and Spatially-Controllable Visual Effects GenerationAlibaba · Proceedings of the AAAI Conference on Artificial Intelligence1 citationMar 14