Residual-Stream Burden Shapes Representation Learning in Diffusion Transformers
cs.CV, cs.LG
Submitted: 2026-09-27
Updated: 2026-10-06
Code: https://github.com/NJU-PCALab/DiP
Terminology
Sources
- DuSPiT: Dual-Branch Sub-Patch Pixel Diffusion Transformer
- Asymmetric Flow Models
- PixelFlow: Pixel-Space Generative Models with Flow
- DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
- Equivalent Flows, Unequal Learning: Clean-Latent Prediction in Transformers
- HyperDiT: Hyper-Connected Transformers for High-Fidelity Pixel-Space Diffusion
- LatentUM: Unleashing the Potential of Interleaved Cross-Modal Reasoning via a Latent-Space Unified Model
- Revisiting Diffusion Model Predictions Through Dimensionality
- Does Sparse Connectivity Improve Generalization? Convolutional Networks Below the Edge of Stability
- Virtual Width Networks
- Improved Baselines with Representation Autoencoders
- Rethinking Cross-Layer Information Routing in Diffusion Transformers
- xHC: Expanded Hyper-Connections
- Frac-Connections: Fractional Extension of Hyper-Connections
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models