KlingAI Research · 18 records

dateidtitleauthorsvenue↑
2026-11-202611.26380EgoCaptioner: Towards Comprehensive and Fine-Grained Egocentric Video CaptioningYuqi Tang et al.arXiv24
2026-11-192611.26595Coda-DaVinci: Reimagining Video Generation with Code2VideoYiyan Ji et al.arXiv32
2026-11-032611.04916Behind Video Generation: Observations and Insights on Multimodal Understanding from KlingKling Team et al.arXiv65
2026-10-222610.29441Sparsity Should Be ALLOTted by Output: Training-Free Allocation of Block-Sparse Attention for Video Diffusion TransformersTianze Xia et al.arXiv21
2026-10-122610.12572DiffusionART: Efficient, Stable and Scalable Reinforcement of Diffusion ModelsZhenglin Wan et al.arXiv103
2026-08-072608.04956ContextMaster: Interactive Multi-Shot Video Creation via Fixed-Budget Sparse Context RoutingXu Guo et al.arXiv5
2026-08-042608.01628Motion Beyond Morphology: Bootstrapping Cross-Category Motion Transfer from Abstract Motion RepresentationsZhixue Fang et al.arXiv16
2026-07-312607.28595Beacon: Knowing When and How to Perform Agentic Visual ReasoningQixun Wang et al.arXiv55
2026-07-312607.28509RefCaptioner: Multi-Reference Image-Grounded Video CaptioningTengfei Liu et al.arXiv28
2026-07-172607.14202KeyFrame-Compass: Towards Comprehensive Evaluation of Keyframe-Conditioned Video GenerationYuqi Tang et al.arXiv41
2026-07-172607.14189MultiRef-Compass: Towards Comprehensive Evaluation of Multi-Reference-to-Audio-Video GenerationXiaohan Zhang et al.arXiv30
2026-06-252606.21661UnityShots: Memory-Driven Multi-Shot Audio-Video Generation with Boundary-Aware GatingJiehui Huang et al.arXiv24
2026-06-152606.13432OmniDirector: General Multi-Shot Camera Cloning without Cross-Paired DataJiwen Liu et al.arXiv54
2026-06-082606.07326AnchorWorld: Embodied Egocentric World Simulation with View-based Evolution CustomizationYu Li et al.arXiv30
2026-06-022606.02564VLMs are Good Teachers for Video Reasoning via Adaptive Test-Time OptimizationJunhao Cheng et al.arXiv25
2026-06-012605.31336DecMem: Towards Minute-Long Consistent World Generation with Decoupled MemoryZhenhao Yang et al.arXiv10
2026-05-272605.26244LongAV-Compass: Towards Unified Evaluation of Minute-Scale Audio-Visual Generation Across T2AV, I2AV, and V2AVTengfei Liu et al.arXiv34
2026-05-222605.22012LatentOmni: Rethinking Omni-Modal Understanding via Unified Audio-Visual Latent ReasoningYifan Dai et al.arXiv44