Temporal-Attention Head Specialization During Video Diffusion Training

arXiv:2609.31654 · cs.CV, cs.AI · Submitted 2026-09-14 · Read on arXiv

cs.CV, cs.AI

Submitted: 2026-09-14

Updated: 2026-09-29

Terminology

• Ablation — 22× in this paper · explained in 1 episode
• Transformer — 21× in this paper · explained in 2 episodes
• Diffusion Model — 10× in this paper · explained in 8 episodes
• Diffusion Models — 9× in this paper · explained in 6 episodes
• Sparsity — 8× in this paper · explained in 4 episodes
• Generative Model — 2× in this paper · explained in 2 episodes
• Generative Models — 2× in this paper · explained in 2 episodes
• In-Context Learning — 2× in this paper · explained in 4 episodes
• Latent Diffusion Model — 2× in this paper · explained in 2 episodes
• Latent Diffusion Models — 2× in this paper · explained in 2 episodes
• Pruning — 2× in this paper · explained in 4 episodes
• Robustness — 2× in this paper · explained in 5 episodes
• Adam — 1× in this paper · explained in 2 episodes
• Aggregation — 1× in this paper · explained in 1 episode
• Attention Mechanism — 1× in this paper · explained in 2 episodes
• Benchmark — 1× in this paper · explained in 2 episodes
• Exchangeability — 1× in this paper · explained in 2 episodes
• Generative AI — 1× in this paper · explained in 2 episodes
• Interpretability — 1× in this paper · explained in 15 episodes
• Joint Attention — 1× in this paper · explained in 2 episodes

Sources

• Active-Dormant Attention Heads: Mechanistically Demystifying Extreme-Token Phenomena in LLMs
• LTX-Video: Realtime Video Latent Diffusion
• HunyuanVideo: A Systematic Framework For Large Video Generative Models
• Understanding Attention Mechanism in Video Diffusion Models
• VDT: General-purpose Video Diffusion Transformers via Mask Modeling
• Emergent Temporal Correspondences from Video Diffusion Transformers
• Progress measures for grokking via mechanistic interpretability
• In-context Learning and Induction Heads
• Wan: Open and Advanced Large-Scale Video Generative Models
• Analysis of Attention in Video Diffusion Transformers
• When Do Attention Circuits Form? Developmental Trajectories of Capability and Attention-Sink Emergence Across Three 1B-ClassArchitectures
• Open-Sora: Democratizing Efficient Video Production for All

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