17 papers
VA-Judger: Reward Modeling from Human Preference Feedback for Joint Video-Audio Generation
Yinming Huang, Shuyuan Tu, Xi Yan +6
Using reinforcement learning to post-train joint video-audio generation models requires a reward signal. Existing methods construct this reward by combining metrics for individual…
CopyCat: Improving Fine-Grained Subject Consistency in Subject-to-Image Models within Seconds
Peng Zheng, Ruiqi Liu, Rui Ma +1
Recent subject-to-image models have achieved impressive progress in personalized image generation, yet they still struggle to preserve fine-grained subject-specific details. A majo…
ARM: An AutoRegressive Large Multimodal Model with Unified Discrete Representations
Junke Wang, Xiao Wang, Jiacheng Pan +16
This paper introduces ARM, a discrete representation-based AutoRegressive Model that unifies image understanding, generation, and editing within a next-token prediction framework.…
IDEAL: In-DEpth ALignment Makes A Discrete Representation AutoEncoder
Yitong Chen, Zijie Diao, Junke Wang +5
Built on pretrained vision foundation models (VFMs), representation autoencoders (RAEs) have recently emerged as a promising approach for constructing semantically rich latent spac…
OmniGen-AR: AutoRegressive Any-to-Image Generation
Junke Wang, Xun Wang, Qiushan Guo +4
Autoregressive (AR) models have demonstrated strong potential in visual generation, offering superior performance with simple architectures and optimization objectives. However, ex…
DisCo: World Models with Discrete Camera Motion Control
Hongrui Huang, Junke Wang, Quanhao Li +2
Controllable video world models target interactive world exploration, where models must faithfully execute explicit action commands while preserving visual quality and temporal coh…
Channel-wise Vector Quantization
Wei Song, Tianhang Wang, Yitong Chen +5
We present Channel-wise Vector Quantization (CVQ), a novel image tokenization paradigm that replaces patch-wise tokens with channel-wise tokens. Unlike conventional vector quantiza…
Baton: Explicit Semantic Blueprints for Joint Video-Audio Generation
Shuyuan Tu, Qi Tian, Zihan Yang +9
Current open-source diffusion models struggle to generate stable and synchronized audio-visual content, particularly in scenarios demanding complex semantic reasoning. The root cau…
DecQ: Detail-Condensing Queries for Enhanced Reconstruction and Generation in Representation Autoencoders
Tianhang Wang, Yitong Chen, Wei Song +3
Representation Autoencoders (RAEs) leverage frozen vision foundation models (VFMs) as tokenizer encoders, providing robust high-level representations that facilitate fast convergen…
Resolving Representation Ambiguity in Feedforward Novel View Synthesis Transformer via Semantic-Spatial Decoupling
Yihang Wu, Yihang Sun, Shaofeng Zhang +4
Transformer-based models have advanced feedforward novel view synthesis (NVS). Current architectures such as GS-LRM and LVSM mix semantic information (e.g., RGB) and spatial inform…
Bench2Drive-Robust: Benchmarking Closed-Loop Autonomous Driving under Deployment Perturbations
Zhiyuan Zhang, Zhenghao Jin, Yanlun Peng +8
Robustness is a critical requirement for deploying autonomous driving systems in the real world. Existing robustness benchmarks for autonomous driving have made important progress…
FluxMem: Adaptive Hierarchical Memory for Streaming Video Understanding
Yiweng Xie, Bo He, Junke Wang +3
This paper presents FluxMem, a training-free framework for efficient streaming video understanding. FluxMem adaptively compresses redundant visual memory through a hierarchical, tw…
Preference Score Distillation: Leveraging 2D Rewards to Align Text-to-3D Generation with Human Preference
Jiaqi Leng, Shuyuan Tu, Haidong Cao +4
Human preference alignment presents a critical yet underexplored challenge for diffusion models in text-to-3D generation. Existing solutions typically require task-specific fine-tu…
ArcFlow: Unleashing 2-Step Text-to-Image Generation via High-Precision Non-Linear Flow Distillation
Zihan Yang, Shuyuan Tu, Licheng Zhang +3
Diffusion models have achieved remarkable generation quality, but they suffer from significant inference cost due to their reliance on multiple sequential denoising steps, motivati…
TempoMaster: Efficient Long Video Generation via Next-Frame-Rate Prediction
Yukuo Ma, Cong Liu, Junke Wang +5
We present TempoMaster, a novel framework that formulates long video generation as next-frame-rate prediction. Specifically, we first generate a low-frame-rate clip that serves as…
Rethinking Discrete Tokens: Treating Them as Conditions for Continuous Autoregressive Image Synthesis
Peng Zheng, Junke Wang, Yi Chang +3
Recent advances in large language models (LLMs) have spurred interests in encoding images as discrete tokens and leveraging autoregressive (AR) frameworks for visual generation. Ho…
FreeLoRA: Enabling Training-Free LoRA Fusion for Autoregressive Multi-Subject Personalization
Peng Zheng, Ye Wang, Rui Ma +1
Subject-driven image generation plays a crucial role in applications such as virtual try-on and poster design. Existing approaches typically fine-tune pretrained generative models…