15 papers
Cooperative Multi-Agent Vision-Language-Action Models via Reinforced Fine Tuning
Ruixiao Xu, Wong Lik Hang Kenny, Zhiqian Liu +12
We study reinforcement learning (RL) methods for cooperative multi-agent Vision-Language-Action (VLA) models. This problem is challenging because VLAs are pretrained on large-scale…
LaCache: Exact Caching and Precision-Adaptive Inference for Diffusion Large Language Models
Xingru Chen, Zelang Liang, Yongjia Ma +4
Diffusion-based Large Language Models(DLLMs) enable parallel generation via Semi-Autoregressive (SAR) decoding in text generation. However, current methods suffer from severe opera…
CogPortrait: Fine-Grained Eye-Region Control in Portrait Animation via Hierarchical Agent Planning
He Feng, Yongjia Ma, Donglin Di +2
Portrait animation methods have achieved substantial visual quality and lip synchronization, but fine-grained manipulation of the eye region still faces a trade-off between input g…
EverybodyDance: Bipartite Graph-Based Identity Correspondence for Multi-Character Animation
Haotian Ling, Zequn Chen, Qiuying Chen +6
Consistent pose-driven character animation has achieved remarkable progress in single-character scenarios. However, extending these advances to multi-character settings is non-triv…
MoCA: Identity-Preserving Text-to-Video Generation via Mixture of Cross Attention
Qi Xie, Yongjia Ma, Donglin Di +2
Achieving ID-preserving text-to-video (T2V) generation remains challenging despite recent advances in diffusion-based models. Existing approaches often fail to capture fine-grained…
DiTalker: A Unified DiT-based Framework for High-Quality and Speaking Styles Controllable Portrait Animation
He Feng, Yongjia Ma, Donglin Di +3
Portrait animation aims to synthesize talking videos from a static reference face, conditioned on audio and style frame cues (e.g., emotion and head poses), while ensuring precise…
QR-LoRA: Efficient and Disentangled Fine-tuning via QR Decomposition for Customized Generation
Jiahui Yang, Yongjia Ma, Donglin Di +6
Existing text-to-image models often rely on parameter fine-tuning techniques such as Low-Rank Adaptation (LoRA) to customize visual attributes. However, when combining multiple LoR…
Adams Bashforth Moulton Solver for Inversion and Editing in Rectified Flow
Yongjia Ma, Donglin Di, Xuan Liu +4
Rectified flow models have achieved remarkable performance in image and video generation tasks. However, existing numerical solvers face a trade-off between fast sampling and high…
UniCP: A Unified Caching and Pruning Framework for Efficient Video Generation
Wenzhang Sun, Qirui Hou, Donglin Di +3
Diffusion Transformers (DiT) excel in video generation but encounter significant computational challenges due to the quadratic complexity of attention. Notably, attention differenc…
Tuning-Free Long Video Generation via Global-Local Collaborative Diffusion
Yongjia Ma, Junlin Chen, Donglin Di +6
Creating high-fidelity, coherent long videos is a sought-after aspiration. While recent video diffusion models have shown promising potential, they still grapple with spatiotempora…
TV-3DG: Mastering Text-to-3D Customized Generation with Visual Prompt
Jiahui Yang, Donglin Di, Baorui Ma +8
In recent years, advancements in generative models have significantly expanded the capabilities of text-to-3D generation. Many approaches rely on Score Distillation Sampling (SDS)…
DH-FaceVid-1K: A Large-Scale High-Quality Dataset for Face Video Generation
Donglin Di, He Feng, Wenzhang Sun +6
Human-centric generative models are becoming increasingly popular, giving rise to various innovative tools and applications, such as talking face videos conditioned on text or audi…
TrAME: Trajectory-Anchored Multi-View Editing for Text-Guided 3D Gaussian Splatting Manipulation
Chaofan Luo, Donglin Di, Xun Yang +4
Despite significant strides in the field of 3D scene editing, current methods encounter substantial challenge, particularly in preserving 3D consistency in multi-view editing proce…
Real Face Video Animation Platform
Xiaokai Chen, Xuan Liu, Donglin Di +3
In recent years, facial video generation models have gained popularity. However, these models often lack expressive power when dealing with exaggerated anime-style faces due to the…
One-Shot Pose-Driving Face Animation Platform
He Feng, Donglin Di, Yongjia Ma +2
The objective of face animation is to generate dynamic and expressive talking head videos from a single reference face, utilizing driving conditions derived from either video or au…