Research map / Video / Video-language understanding
KDA: Knowledge Diffusion Alignment with Enhanced Context for Video Temporal Grounding
Ran Ran, Jiwei Wei, Shiyuan He, Zeyu Ma, Chaoning Zhang, Ning Xie, Yang Yang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add MKDM diffusion moment knowledge (ablation) | KDA baseline · QVHighlights val (SF+C) | R1@0.5: 64.64 -> 66.05 | Improved |
| add MVKE multi-layer knowledge extraction (ablation) | KDA with MKDM and SED · QVHighlights val (SF+C) | mAP avg: 46.08 -> 47.41 | Improved |
| add spans-enhanced decoder (ablation) | KDA with MKDM and MVKE · QVHighlights val (SF+C) | mAP avg: 45.79 -> 47.41 | Improved |
| use BRDM instead of DDPM (ablation) | KDA diffusion · QVHighlights val (SF+C) | mAP avg: 45.53 -> 47.41 | Improved |
| use BRDM (4 steps) instead of DDIM (4 steps) (ablation) | KDA diffusion · QVHighlights val (SF+C) | mAP avg: 44.78 -> 47.41 | Improved |
| add diffusion model with 4 steps (ablation) | KDA · QVHighlights val (SF+C) | mAP avg: 44.86 -> 47.41 | Improved |
| increase diffusion steps from 4 to 8 (ablation) | KDA · QVHighlights val (SF+C) | mAP avg: 47.41 -> 47.30; time 9.3s -> 11.5s | No clear effect |
| use adaptive spans and query contents instead of spans only (ablation) | KDA decoder · QVHighlights val (SF+C) | mAP avg: 47.41 -> 44.76 | Hurt |
| use static spans and query instead of SED (ablation) | KDA decoder · QVHighlights val (SF+C) | mAP avg: 47.41 -> 45.86 | Hurt |
| KDA vs UVCOM on TACoS | KDA · TACoS (SF+C) | R1@0.5: 36.39 -> 40.13 | Improved |
| KDA vs TR-DETR on Charades-STA VGG | KDA · Charades-STA (VGG) | R1@0.7: 30.81 -> 34.50 | Improved |
Closest papers by meaning
- Knowledge Diffusion-Based Adaptive Alignment With Hierarchical Context for Video Temporal Grounding (TPAMI 2026 · similarity 0.97)
- SE-DETR: Explicit Semantic Exploration for Generalizability and Distinguishability in Video Temporal Grounding (ECCV 2026 · similarity 0.88)
- Generative Video Diffusion for Unseen Novel Semantic Video Moment Retrieval (AAAI 2025 · similarity 0.87)
- Uncertainty-quantified Rollout Policy Adaptation for Unlabelled Cross-domain Video Temporal Grounding (NeurIPS 2025 · similarity 0.87)
- Inference-Time Text-to-Video Alignment with Diffusion Latent Beam Search (NeurIPS 2025 · similarity 0.86)
- TOPA: Extending Large Language Models for Video Understanding via Text-Only Pre-Alignment (NeurIPS 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
