TCoVP: Spatiotemporally Grounded Video Chain-of-Thought Improves Reasoning Segmentation
Junwei Wen · hf · 2026-10-20
Researchers from HIT Shenzhen, Pengcheng Lab and UC Merced propose TCoVP (Temporal Chain of Video Primitives), addressing the 'video reference gap' where text-only CoT explains a target choice without explicit visual references.
- Organizes query-relevant observations into a chronological reasoning chain, each primitive linking an entity to temporal support and timestamped spatial evidence
- Selected primitives provide box prompts and temporal support for video mask prediction
- Built via TCoVP-SFT-15k supervised fine-tuning and TCoVP-RL-13k chain-aligned RL, improving reasoning and masks across five VRS benchmarks
More from Multimodal
- MOSS Image with 8-step Turbo LoRA tested: 10x faster with minimal quality loss — Aggressive_Collar135 · 2026-10-20
- User turns their AI muse Maki into a punky ska song with Suno V6 — thederbiedone · 2026-10-20
- Grok Imagine parody short 'Wendy' shows off AI-made TV skits — Old_Put1306 · 2026-10-20
- AI-generated "Back to the Future" sequel clip goes viral for Halloween — mitchdeg · 2026-10-20
- Midjourney v8.2 style ref 1631327937 yields a dreamy pastel look — azed_ai · 2026-10-20
- Google Flow hosts hybrid production workshop pairing live-action with AI filmmaking workflows — lmoroney · 2026-10-20