Fudan University · 25 records

dateidtitleauthorsvenue↑
2026-11-052611.06425D3D: Decoupled Diffusion Drafting for Asynchronous Speculative DecodingBaoyou Chen et al.arXiv17
2026-11-042611.04303FlowTranslator: Training-Free In-Image Machine Translation with Flow-Based EditingZhixian Li et al.arXiv11
2026-10-232610.28416Code2Video-Compass: A Comprehensive Benchmark for General-Purpose Code-Based Video Generation across T2V, I2V, and V2VZhenghan Chen et al.arXiv34
2026-10-222610.27592Real In, Real Out: What If We Only Use Real Data for Scene Text Editing?Xingsong Ye et al.arXiv20
2026-10-192610.09872LiveMACE: Process-Aware Evaluation of LLM Agent Capabilities in Evolving MarketsJun Zhao et al.arXiv52
2026-10-192610.14746BridgeWAM: Bridging Video and Action Experts via Latent Bridge QueriesJiaxin Fang et al.arXiv44
2026-10-162610.22186EVA-Cap: Optimizing Audiovisual Video Captioning via Event-Centric AlignmentJiapeng Shi et al.arXiv4
2026-10-152610.19366Seeing, Maintaining, and Learning: The Evolution of Multimodal Agent MemoryMinghao Guo et al.arXiv52
2026-10-012609.37569Decompose Radicals, Then Reward: Fine-Grained Inspection for Accurate Chinese Text RenderingYazhen Xie et al.arXiv7
2026-09-302609.36965Chinese-Jev: Bringing System One Model to Chinese-Language TasksZexiao Wang et al.arXiv16
2026-09-292609.35134VideoPhysEdit: Physical Counterfactual Video Editing via Rigid-Body Physical Scene ReconstructionConghan Yue et al.arXiv16
2026-09-232609.24058All-in-One Multilingual Scene Text Recognition with Script-aware Mixture-of-ExpertsXingsong Ye et al.arXiv53
2026-09-092609.06758Agentic Visual Generation: From Generative Models to Agentic ControlYinming Huang et al.arXiv21
2026-09-092609.08404Environments as Scaffold: Enriching Feedback to Bootstrap Self-Evolving Agents in Long-Horizon TasksHongbang Yuan et al.arXiv10
2026-09-022608.30478Agents in the Large: Perception-Centered Architecture for Persistent AgentsShihan Dou et al.arXiv12
2026-08-282608.26993Aphanta: Diagnosing Task-Aligned Image-Edited Intermediates for Multimodal ReasoningHengyuan Xu et al.arXiv3
2026-07-162607.05910PolicyShiftGuard: Benchmarking and Improving Policy-Adaptive Image GuardrailsMingyang Song et al.arXiv3
2026-06-252606.24484Advancing WordArt-Oriented Scene Text Recognition: Datasets and MethodsXingsong Ye et al.arXiv6
2026-06-192606.20506FreeStyle: Free Control of Style-Content Dual-Reference Generation from Community LoRA MiningJinghong Lan et al.arXiv26
2026-06-112606.13674RepWAM: World Action Modeling with Representation Visual-Action TokenizersJunke Wang et al.arXiv6
2026-05-292605.30265LoMo: Local Modality Substitution for Deeper Vision-Language FusionFeng Han et al.arXiv20
2026-05-292605.22189Learning A Unified Risk Map for Autonomous Driving in Partially Observable EnvironmentsJie Jia et al.arXiv4
2026-05-282605.28293ProRL: Effective Reinforcement Learning for Proactive Recommendation via Rectified Policy Gradient EstimationHongru Hou et al.arXiv81
2026-05-282605.28421DenoiseRL: Bootstrapping Reasoning Models to Recover from Noisy PrefixesCaijun Xu et al.arXiv45
2026-05-122605.07465SEIF: Self-Evolving Reinforcement Learning for Instruction FollowingQingyu Ren et al.arXiv9