11 papers
NesTok: Nested Self-Aligned 1D Tokenizer for Autoregressive Image Generation
Jiawei Zhang, Shuhao Liu, Rong Huang +4
One-dimensional (1D) variable-length visual tokenizers enable adaptive compression by varying the number of tokens, allowing downstream autoregressive (AR) models to flexibly trade…
SRC-Flow: Compact Semantic Representations Enable Normalizing Flows for Image Generation
Longtao Jiang, Jianmin Bao, Zhendong Wang +4
Normalizing flows (NFs) provide exact likelihoods and deterministic invertible sampling, but have historically lagged behind diffusion models for large-scale image generation. We i…
LatentPilot: Scene-Aware Vision-and-Language Navigation by Dreaming Ahead with Latent Visual Reasoning
Haihong Hao, Lei Chen, Mingfei Han +5
Existing vision-and-language navigation (VLN) models primarily reason over past and current visual observations, while largely ignoring the future visual dynamics induced by action…
Beyond Dense Futures: World Models as Structured Planners for Robotic Manipulation
Minghao Jin, Mozheng Liao, Mingfei Han +2
Recent world-model-based Vision-Language-Action (VLA) architectures have improved robotic manipulation through predictive visual foresight. However, dense future prediction introdu…
User-Feedback-Driven Adaptation for Vision-and-Language Navigation
Yongqiang Yu, Xuhui Li, Hazza Mahmood +6
Real-world deployment of Vision-and-Language Navigation (VLN) agents is constrained by the scarcity of reliable supervision after offline training. While recent adaptation methods…
Efficient Training for Human Video Generation with Entropy-Guided Prioritized Progressive Learning
Changlin Li, Jiawei Zhang, Shuhao Liu +4
Human video generation has advanced rapidly with the development of diffusion models, but the high computational cost and substantial memory consumption associated with training th…
Which Layer Causes Distribution Deviation? Entropy-Guided Adaptive Pruning for Diffusion and Flow Models
Changlin Li, Jiawei Zhang, Zeyi Shi +3
Large-scale vision generative models, including diffusion and flow models, have demonstrated remarkable performance in visual generation tasks. However, transferring these pre-trai…
Self-Consistency as a Free Lunch: Reducing Hallucinations in Vision-Language Models via Self-Reflection
Mingfei Han, Haihong Hao, Jinxing Zhou +5
Vision-language models often hallucinate details, generating non-existent objects or inaccurate attributes that compromise output reliability. Existing methods typically address th…
Token Painter: Training-Free Text-Guided Image Inpainting via Mask Autoregressive Models
Longtao Jiang, Jie Huang, Mingfei Han +5
Text-guided image inpainting aims to inpaint masked image regions based on a textual prompt while preserving the background. Although diffusion-based methods have become dominant,…
Mettle: Meta-Token Learning for Memory-Efficient Audio-Visual Adaptation
Jinxing Zhou, Zhihui Li, Yongqiang Yu +7
We present \textbf{Met}a-\textbf{T}oken \textbf{Le}arning (Mettle), a simple and memory-efficient method for adapting large-scale pretrained transformer models to downstream audio-…
CoNav: Collaborative Cross-Modal Reasoning for Embodied Navigation
Haihong Hao, Mingfei Han, Changlin Li +2
Embodied navigation demands comprehensive scene understanding and precise spatial reasoning. While image-text models excel at interpreting pixel-level color and lighting cues, 3D-t…