3 papers
cs.CV2026
When Attention Collapses: Stage-Aware Visual Token Pruning from Structure to Semantics
Jiahui Wang, Kai Zhang, Mai Han +1
Vision-Language Models (VLMs) have demonstrated remarkable capabilities but suffer from significant computational overhead during inference. While visual token pruning offers a pro…
cs.CV2025
Both Semantics and Reconstruction Matter: Making Representation Encoders Ready for Text-to-Image Generation and Editing
Shilong Zhang, He Zhang, Zhifei Zhang +11
Modern Latent Diffusion Models (LDMs) typically operate in low-level Variational Autoencoder (VAE) latent spaces that are primarily optimized for pixel-level reconstruction. To uni…
cs.AI2025
PhyX: Does Your Model Have the "Wits" for Physical Reasoning?
Hui Shen, Taiqiang Wu, Qi Han +16
Existing benchmarks fail to capture a crucial aspect of intelligence: physical reasoning, the integrated ability to combine domain knowledge, symbolic reasoning, and understanding…