Research map / Video / Video-language understanding
Memory-Guided Semantic Learning Network for Temporal Sentence Grounding
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| replace the strongest listed baseline CBLN with MGSL-Net | temporal sentence grounding · ActivityNet Caption | R@1, IoU=0.7: 27.60 -> 31.42 | Helped |
| replace the strongest listed baseline CBLN with MGSL-Net | temporal sentence grounding · TACoS | R@1, IoU=0.5: 27.65 -> 32.27 | Helped |
| replace the strongest listed baseline CBLN with MGSL-Net | temporal sentence grounding · Charades-STA | R@1, IoU=0.7: 38.22 -> 41.03 | Helped |
| add domain-specific memory networks to the baseline (ablation) | MGSL-Net · ActivityNet Caption | R@1, IoU=0.5: 43.65 -> 49.24 | Helped |
| add heterogeneous attention to the baseline (ablation) | MGSL-Net · ActivityNet Caption | R@1, IoU=0.5: 43.65 -> 46.19 | Helped |
| add heterogeneous attention to the memory-only variant (ablation) | MGSL-Net · ActivityNet Caption | R@1, IoU=0.5: 49.24 -> 51.87 | Helped |
| use both aligned pair features in the video-domain memory (ablation) | MGSL-Net memory · ActivityNet Caption | R@1, IoU=0.5: 47.64 -> 48.88 | Helped |
| share memory slots for the paired video-domain features (ablation) | MGSL-Net memory · ActivityNet Caption | R@1, IoU=0.5: 48.88 -> 49.97 | Helped |
| add query-domain memory features to the video-domain memory configuration (ablation) | MGSL-Net memory · ActivityNet Caption | R@1, IoU=0.5: 49.97 -> 50.56 | Helped |
| add aligned video feature V-hat to query-domain memory (ablation) | MGSL-Net memory · ActivityNet Caption | R@1, IoU=0.5: 50.56 -> 51.20 | Helped |
| add shared memory slots in both domains (ablation) | MGSL-Net memory · ActivityNet Caption | R@1, IoU=0.5: 51.20 -> 51.87 | Helped |
| add calibration to inter-attention (ablation) | heterogeneous attention · ActivityNet Caption | R@1, IoU=0.5: 49.24 -> 50.33 | Helped |
Closest papers by meaning
- Multi-Pair Temporal Sentence Grounding via Multi-Thread Knowledge Transfer Network (AAAI 2025 · similarity 0.91)
- Temporal Sentence Grounding with Relevance Feedback in Videos (NeurIPS 2024 · similarity 0.89)
- Empower Words: DualGround for Structured Phrase and Sentence-Level Temporal Grounding (NeurIPS 2025 · similarity 0.88)
- Universal Video Temporal Grounding with Generative Multi-modal Large Language Models (NeurIPS 2025 · similarity 0.86)
- VideoLucy: Deep Memory Backtracking for Long Video Understanding (NeurIPS 2025 · similarity 0.86)
- Seq2Time: Sequential Knowledge Transfer for Video LLM Temporal Grounding (CVPR 2025 · 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.