Research map / Video / Video-language understanding
Quantifying and Narrowing the Unknown: Interactive Text-to-Video Retrieval via Uncertainty Minimization
Bingqing Zhang, Zhuo Cao, Heming Du, Yang Li, Xue Li, Jiajun Liu, Sen Wang
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 MUS (mapping uncertainty score) to TAS (ablation) | UMIVR · MSR-VTT-1k (round 3) | Hit@1: 67.0 -> 68.6 | Improved |
| add TQFS (temporal quality-based frame sampler) to TAS + MUS (ablation) | UMIVR · MSR-VTT-1k (round 3) | Hit@1: 68.6 -> 68.9 | Improved |
| add MUS to TAS (ablation) | UMIVR · MSR-VTT-1k (round 5) | R@1: 63.4 -> 64.2 | Improved |
| add TQFS to TAS + MUS (ablation) | UMIVR · MSR-VTT-1k (round 5) | R@1: 64.2 -> 65.0 | Improved |
| add TQFS to TAS + MUS (ablation) | UMIVR · MSR-VTT-1k (round 3) | R@1: 62.1 -> 61.3 | Hurt |
| use TAS/MUS thresholds (0.5, 0.2) instead of (0.4, 0.1) (ablation) | UMIVR · MSR-VTT-1k (round 5) | BRI: 0.73 -> 0.67 | Improved |
| add TQFS plug-in to Xpool (ablation) | Xpool · MSR-VTT-1kA text-to-video | R@10: 80.2 -> 80.9 | Improved |
| add TQFS plug-in to Xpool (ablation) | Xpool · MSR-VTT-1kA video-to-text | R@10: 83.2 -> 82.7 | Hurt |
| add TQFS plug-in to VideoCLIP (ablation) | VideoCLIP · MSR-VTT-1kA text-to-video | R@10: 71.0 -> 72.4 | Improved |
| use 8-bit instead of 4-bit quantization (ablation) | UMIVR · MSR-VTT-1k (round 5) | R@1: 65.0 -> 65.1 | No clear effect |
| remove quantization (full precision) (ablation) | UMIVR · MSR-VTT-1k (round 5) | R@1: 65.0 -> 67.1 | Improved |
| remove quantization (full precision) (ablation) | UMIVR · MSR-VTT-1k (GPU memory) | GPU memory (MB): 10210 -> 19297 | Hurt |
Closest papers by meaning
- Ambiguity-Restrained Text-Video Representation Learning for Partially Relevant Video Retrieval (AAAI 2025 · similarity 0.89)
- Bridging Information Asymmetry in Text-video Retrieval: A Data-centric Approach (ICLR 2025 · similarity 0.88)
- Text Proxy: Decomposing Retrieval from a 1-to-N Relationship into N 1-to-1 Relationships for Text-Video Retrieval (AAAI 2025 · similarity 0.88)
- Mitigating Semantic Collapse in Partially Relevant Video Retrieval (NeurIPS 2025 · similarity 0.87)
- Hierarchical Multi-Source Uncertainty Aggregation for Interactive Video Captioning (AAAI 2025 · similarity 0.87)
- Video-ColBERT: Contextualized Late Interaction for Text-to-Video Retrieval (CVPR 2025 · similarity 0.87)
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.
