Xiaoqin Fu

Erase but Preserve: Controllable Removal of Copyrighted Animation Characters via Optimized Semantic Anchors

Qiao Li, Xiaoqin Fu, Wangjia Yu, Runze He et autres

The exceptional generation capabilities of text-to-image diffusion models have raised copyright concerns, particularly the unauthorized reproduction of animation characters. Existing concept erasure methods fall short for animation character erasure: model modification methods struggle to identify suitable anchors for diverse, highly distinctive characters; …

Semantic Steering for Controllable Generation: Tuning-Free Concept Erasure in Multimodal Diffusion Transformers

Qiao Li, Xiaoqin Fu, Yuanshu Zhao, Qipeng Wang et autres

Multimodal Diffusion Transformers (MM-DiTs) have demonstrated remarkable text-to-image generation performance, surpassing traditional U-Net-based diffusion models. Nevertheless, their powerful generative capabilities also raise significant safety concerns, as they may generate sensitive or inappropriate content. While existing concept erasure methods aim to mitigate such …

Towards Memory-Efficient Autoregressive Video Generation via Instance-Specific Parametric Absorption

Xiaoqin Fu, Jia Li, Yiming Hu, Yong Wang et autres

Autoregressive (AR) streaming models have emerged as a powerful paradigm for long video generation. However, the linearly growing Key-Value (KV) cache poses a significant bottleneck, leading to memory overload and degraded inference throughput. A common compression method is to drop redundant KV …

In-Context Model Predictive Generation: Open-Vocabulary Motion Synthesis from Language Models to Physics

Xiaoqin Fu, Junfan Lin, Yang Liu, Yaowei Wang et autres

Synthesizing human motion from textual descriptions is essential for immersive digital applications, yet existing methods face a persistent trade-off between semantic fidelity and physical realism. Large language model (LLM)-based approaches can interpret diverse open-vocabulary instructions and compose high-level action plans, but they …

In-Context Model Predictive Generation: Open-Vocabulary Motion Synthesis from Language Models to Physics

Xiaoqin Fu, Junfan Lin, Yang Liu, Yaowei Wang et autres

Synthesizing human motion from textual descriptions is essential for immersive digital applications, yet existing methods face a persistent trade-off between semantic fidelity and physical realism. Large language model (LLM)-based approaches can interpret diverse open-vocabulary instructions and compose high-level action plans, but they …

Latent DPO for Concept Erasure in Text-To-Video Diffusion Models

Baisen Wang, Xiaoqin Fu, Runze He, Qi Li et autres

We address concept erasure in text-to-video models, targeting harmful concepts like copyrighted identities, celebrities, while preserving visual quality and temporal coherence. We cast erasure as latent preference learning: a decoding-free adaptation of Direct Preference Optimization (DPO) trains the denoiser, under matched noise …

cn, sg (code pays fourni par la source)

Train Short, Inference Long: Training-free Horizon Extension for Autoregressive Video Generation

Jia Li, Xiaoqin Fu, Xurui Peng, Weifeng Chen et autres

Autoregressive video diffusion models have emerged as a scalable paradigm for long video generation. However, they often suffer from severe extrapolation failure, where rapid error accumulation leads to significant temporal degradation when extending beyond training horizons. We identify that this failure primarily …

Train Short, Inference Long: Training-free Horizon Extension for Autoregressive Video Generation

Jia Li, Xiaoqin Fu, Xurui Peng, Weifeng Chen et autres

Autoregressive video diffusion models have emerged as a scalable paradigm for long video generation. However, they often suffer from severe extrapolation failure, where rapid error accumulation leads to significant temporal degradation when extending beyond training horizons. We identify that this failure primarily …

Towards Memory-Efficient Autoregressive Video Generation via Instance-Specific Parametric Absorption

Xiaoqin Fu, Jia Li, Yiming Hu, Yong Wang et autres

cn, hk (code pays fourni par la source)

Neuroprotective potential of Rhus coriaria : Insights into acetylcholinesterase inhibition and amyloid-β modulation

Adeel Ahmed, Hamid Khan, Yin Shen, Shishi Xu et autres

Sumac (Rhus coriaria) is a spice and a medicinal plant that has been indicated to exert favorable effects in the management of different diseases, and it also possesses high anti-inflammatory and antioxidant properties. A plant known to be rich in bioactive molecules, …

cn (code pays fourni par la source)

Adaptive Test-Time Semantic Debiasing for AI-Generated Image Detection

Yu Cai, Jiahe Tian, Xiaoqin Fu, Jiao Dai et autres

AI-generated image detectors have historically concentrated on generalization across generative models, often overlooking the critical challenge of cross-semantic generalizability. This limitation constrains the adaptability of detectors to new semantic content in real-world settings. We propose Adaptive Test-Time Semantic Debiasing (ATTSD), a zero-shot …

us, cn (code pays fourni par la source)

TCFG: Truncated Classifier-Free Guidance for Efficient and Scalable Text-to-Image Acceleration

Xiaoqin Fu, J J Li

cn, hk (code pays fourni par la source)