Nankai Univ. Releases Thinking on Shots: Agentic Multi-Shot Video Editing
NankaiUniversity · hf · 2026-08-28
Nankai University released Thinking on Shots, an agentic framework combining LLMs and VLMs. It enables consistent, multi-instruction editing of long multi-shot videos while preserving spatiotemporal structure.
More from Multimodal
- ChatGPT Adds 'References' for Consistent Image Generation — btibor91 · 2026-08-28
- 15-second AI fashion film treats typography as choreography — CloudwiseAIJP · 2026-08-28
- 30-second photorealistic Japanese horror short made with Seedance 2.5 — SimplyAnnisa · 2026-08-28
- Rainy bus-shelter portrait: a full Midjourney prompt worth stealing — tisch_eins · 2026-08-28
- Seadance 2.5 Test: 30s Video from Single Prompt and 3 References for $4 — justin_hart · 2026-08-28
- Lux3D generates 3D assets from a single image — SarahAnnabels · 2026-08-28