11 citations · 24 across the 13 of their papers we have counts for
14 papers
TwinPrune: Density-Aware Two-Phase Token Pruning for Vision-Language Models
Kai Liu, Anqi Li, Junxian Li +4
We propose a unified spatial-density signal to score token importance for both inference phases: bipartite merge for prefill and KV eviction for decode, yielding measurable gains i…
Illuminating Unified Multimodal Model for Free-form Interleaved Text-Image Generation
Chonghuinan Wang, Zhikai Chen, Chunwei Wang +9
The advancement of generative AI models capable of producing text and image marks a critical step forward in the realm of multimodal intelligence, particularly for tasks involving…
ShotCrop: Cropping Human-Centric Images into Cinematic Triple-Shot Compositions
Dehong Kong, Lina Lei, Lingtao Zheng +10
Prior work on aesthetic composition typically produces a single aesthetically pleasing crop, overlooking the narrative value of composing multiple shots from one scene. In practice…
PermuQuant: Lowering Per-Group Quantization Error by Reordering Channels for Diffusion Models
Yongsen Cheng, Kai Liu, Kaiwen Tao +5
Large-scale visual generative models have achieved remarkable performance. However, their high computational and memory costs make deployment challenging in resource-constrained sc…
Accelerating Rectified Flow Models via Trajectory-Aware Caching
Xiao Liu, Kai Liu, Naiyang Guan +5
Diffusion and rectified flow (RF) models generate high-fidelity images and videos, but their iterative velocity-field evaluations are computationally expensive. Existing caching me…
GTR: Generation-Guided Visual Token Reduction for Separate-Encoder Unified Multimodal Models
Junxian Li, Kai Liu, Zizhong Ding +4
The development of separate-encoder Unified multimodal models (UMMs) comes with a rapidly growing inference cost due to dense visual token processing. In this paper, we focus on un…
HP-Edit: A Human-Preference Post-Training Framework for Image Editing
Fan Li, Chonghuinan Wang, Lina Lei +9
Common image editing tasks typically adopt powerful generative diffusion models as the leading paradigm for real-world content editing. Meanwhile, although reinforcement learning (…
UmniBench: Unified Understand and Generation Model Oriented Omni-dimensional Benchmark
Kai Liu, Leyang Chen, Wenbo Li +5
Unifying multimodal understanding and generation has shown impressive capabilities in cutting-edge proprietary systems. However, evaluations of unified multimodal models (UMMs) rem…
Learning Spatial Adaptation and Temporal Coherence in Diffusion Models for Video Super-Resolution
Zhikai Chen, Fuchen Long, Zhaofan Qiu +4
Diffusion models are just at a tipping point for image super-resolution task. Nevertheless, it is not trivial to capitalize on diffusion models for video super-resolution which nec…
Inter-frame Accelerate Attack against Video Interpolation Models
Junpei Liao, Zhikai Chen, Liang Yi +3
Deep learning based video frame interpolation (VIF) method, aiming to synthesis the intermediate frames to enhance video quality, have been highly developed in the past few years.…
Anti-Forgery: Towards a Stealthy and Robust DeepFake Disruption Attack via Adversarial Perceptual-aware Perturbations
Run Wang, Ziheng Huang, Zhikai Chen +3
DeepFake is becoming a real risk to society and brings potential threats to both individual privacy and political security due to the DeepFaked multimedia are realistic and convinc…
MagDR: Mask-guided Detection and Reconstruction for Defending Deepfakes
Zhikai Chen, Lingxi Xie, Shanmin Pang +2
Deepfakes raised serious concerns on the authenticity of visual contents. Prior works revealed the possibility to disrupt deepfakes by adding adversarial perturbations to the sourc…
Appending Adversarial Frames for Universal Video Attack
Zhikai Chen, Lingxi Xie, Shanmin Pang +2
There have been many efforts in attacking image classification models with adversarial perturbations, but the same topic on video classification has not yet been thoroughly studied…
Unsupervised Single Image Deraining with Self-supervised Constraints
Xin Jin, Zhibo Chen, Jianxin Lin +2
Most existing single image deraining methods require learning supervised models from a large set of paired synthetic training data, which limits their generality, scalability and p…