TT-VidT: Decoupling the Temporal Axis for Efficient Motion-Centric Video Pretraining Paper • 2609.33419 • Published 11 days ago • 19
BootsofLagrangian/laion-highres-aesthetic-recap-qwen3p5-35b-a3b Viewer • Updated 8 days ago • 96.1M • 73
BootsofLagrangian/laion-aesthetics-recap-qwen3p5-35b-a3b Viewer • Updated 8 days ago • 24.3M • 98 • 1
BootsofLagrangian/commoncatalog-cc-by-recap-qwen3p5-35b-a3b Viewer • Updated 8 days ago • 14.6M • 141 • 1
Recaptioned Image Text Collection NeurIPS 2026 E&D Accept (Poster). Caption-only Qwen3.5 recaptions of nine public image corpora, released with a matched-budget audit framework. • 9 items • Updated 8 days ago