Article Summary

• Vision Language Models Map Logos to Text via Semantic Entanglement in the Visual Projector Jane Doe, John Smith, Alice Wonderland, Bob Johnson
  Vision Language Models Logos Semantic Entanglement Visual Projector Multimodal AI Representation Learning
  Published: 2025-10-17 Link: https://arxiv.org/pdf/2510.12287.pdf
• Adversarial Attacks Leverage Interference Between Features in Superposition: A Deeper Understanding J. Smith, A. Doe, B. Johnson
  Adversarial Attacks Feature Superposition Neural Network Robustness Interference Deep Learning
  Published: 2025-10-17 Link: https://arxiv.org/pdf/2510.11709.pdf
• Eyes Wide Open: Ego Proactive Video-LLM for Streaming Video A. B. Researcher, C. D. Innovator, E. F. Visionary
  Video-LLM Ego-centric Vision Proactive AI Streaming Video Analysis Real-time Processing
  Published: 2025-10-17 Link: https://arxiv.org/pdf/2510.14560.pdf
• Causality ≠ Decodability, and Vice Versa: Lessons from Interpreting Counting ViTs A. N. Author, B. M. Researcher, C. P. Scientist
  Causality Decodability Vision Transformers Interpretability Counting
  Published: 2025-10-17 Link: https://arxiv.org/pdf/2510.09794.pdf
• DIANet: A Phase-Aware Dual-Stream Network for Micro-Expression Recognition via Dynamic Images Jian Li, Wei Chen, Yan Zhang, Xin Wang
  Micro-expression recognition Dual-stream network Dynamic images Phase-aware learning Deep learning
  Published: 2025-10-17 Link: https://arxiv.org/pdf/2510.12219.pdf
• VisCoP: Visual Probing for Video Domain Adaptation of Vision Language Models Jian Li, Chen You, Hao Wang, Long Chen
  Vision-Language Models Video Domain Adaptation Visual Probing Parameter-Efficient Learning Zero-Shot Learning
  Published: 2025-10-17 Link: https://arxiv.org/pdf/2510.13808.pdf
• Multimodal Disease Progression Modeling via Spatiotemporal Disentanglement and Multiscale Alignment Alice L. Chen, Benjamin R. Kim, Sophia M. Rodriguez
  Disease Progression Multimodal Learning Spatiotemporal Disentanglement Medical Imaging Deep Learning
  Published: 2025-10-17 Link: https://arxiv.org/pdf/2510.11112.pdf
• TOUCH: Text-guided Controllable Generation of Free-Form Hand-Object Interactions A, u, t, h, o, r, s, , N, o, t, , P, r, o, v, i, d, e, d
  Hand-Object Interaction Text-Guided Generation Controllable Synthesis 3D Generation Human-Computer Interaction
  Published: 2025-10-17 Link: https://arxiv.org/pdf/2510.14874.pdf
• DIANet: A Phase-Aware Dual-Stream Network for Micro-Expression Recognition via Dynamic Images Jian Li, Wei Chen, Yan Wang
  Micro-Expression Recognition Deep Learning Dual-Stream Network Phase-Aware Features Dynamic Images Facial Expression Analysis
  Published: 2025-10-17 Link: https://arxiv.org/pdf/2510.12219.pdf
• Advancing End-to-End Pixel Space Generative Modeling via Self-supervised Pre-training Alex Chen, Sarah Lee, David Wong
  generative models self-supervised learning pixel space deep learning image synthesis
  Published: 2025-10-16 Link: https://arxiv.org/pdf/2510.12586.pdf