12 papers
Improving Generalization Robustness of Multimodal RLVR
Pengfei Zhou, Zhiwei Tang, Xiaopeng Peng +11
Reinforcement Learning with Verifiable Rewards (RLVR) makes Multimodal Large Language Models more accurate, but the gains are brittle: simply paraphrasing a question or changing th…
WorldOlympiad: Can Your World Model Survive a Triathlon?
Yuke Zhao, Wangbo Zhao, Weijie Wang +8
We introduce WorldOlympiad, a benchmark for diagnosing video-based world models across physical faithfulness, geometric consistency, and interaction fidelity. While existing benchm…
Towards Error-Free Long Video Generation
Shuning Chang, Weihua Chen, Jiasheng Tang +8
Recent advances in video generation have made minute-level synthesis possible; however, generating long videos remains challenging due to error accumulation, attribute drift, and t…
K-Forcing: Joint Next-K-Token Decoding via Push-Forward Language Modeling
Zhiwei Tang, Yuanyu He, Yizheng Han +4
Autoregressive (AR) language modeling is the dominant paradigm for text generation, yet its sequential token-by-token decoding makes inference memory-bound and inefficient. Existin…
Few-Step Distillation for Text-to-Image Generation: A Practical Guide
Yifan Pu, Yizeng Han, Zhiwei Tang +4
Diffusion distillation has dramatically accelerated class-conditional image synthesis, but its applicability to open-ended text-to-image (T2I) generation is still unclear. We prese…
BIFE: Better Interaction, Fewer Errors for Minute-Long Video Generation
Zeyu Zhang, Jinyuan Mao, Shuning Chang +5
Long video generation is a critical step toward building realistic world models, requiring both high visual fidelity and long-range interaction consistency. Recent autoregressive d…
RAPID^3: Tri-Level Reinforced Acceleration Policies for Diffusion Transformer
Wangbo Zhao, Yizeng Han, Zhiwei Tang +7
Diffusion Transformers (DiTs) excel at visual generation yet remain hampered by slow sampling. Existing training-free accelerators - step reduction, feature caching, and sparse att…
Flow Along the K-Amplitude for Generative Modeling
Weitao Du, Shuning Chang, Jiasheng Tang +3
In this work, we propose a novel generative learning paradigm, K-Flow, an algorithm that flows along the -amplitude. Here, is a scaling parameter that organizes frequency ba…
DyDiT++: Diffusion Transformers with Timestep and Spatial Dynamics for Efficient Visual Generation
Wangbo Zhao, Yizeng Han, Jiasheng Tang +6
Diffusion Transformer (DiT), an emerging diffusion model for visual generation, has demonstrated superior performance but suffers from substantial computational costs. Our investig…
SparseDiT: Token Sparsification for Efficient Diffusion Transformer
Shuning Chang, Pichao Wang, Jiasheng Tang +2
Diffusion Transformers (DiT) are renowned for their impressive generative performance; however, they are significantly constrained by considerable computational costs due to the qu…
Dynamic Diffusion Transformer
Wangbo Zhao, Yizeng Han, Jiasheng Tang +5
Diffusion Transformer (DiT), an emerging diffusion model for image generation, has demonstrated superior performance but suffers from substantial computational costs. Our investiga…
Inference-Time Alignment of Diffusion Models with Direct Noise Optimization
Zhiwei Tang, Jiangweizhi Peng, Jiasheng Tang +3
In this work, we focus on the alignment problem of diffusion models with a continuous reward function, which represents specific objectives for downstream tasks, such as increasing…