Research map / Video / Video-language understanding
Towards Neuro-Symbolic Video Understanding
Minkyu Choi, Harsh Goel, Mohammad Omama, Yunhao Yang, Sahil Shah, Sandeep Chinchali
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 |
|---|---|---|---|
| use a larger perception model (YOLOv8n to YOLOv8x) (ablation) | NSVS-TL · Waymo-TLV, A until B | F1: 0.20 -> 0.48 | Improved |
| use Masked R-CNN instead of YOLOv8x as perception model (ablation) | NSVS-TL · Waymo-TLV, (A and B) until C | F1: 0.38 -> 0.80 | Improved |
| use Masked R-CNN instead of YOLOv8x as perception model (ablation) | NSVS-TL · NuScenes-TLV, A until B | F1: 0.32 -> 0.64 | Improved |
| use Masked R-CNN instead of YOLOv8x as perception model (ablation) | NSVS-TL · COCO-TLV + ImageNet-TLV, (A and B) until C | F1: 0.80 -> 0.74 | Hurt |
| use CLIP-ViT-B-32 as perception model for conjunction queries (ablation) | NSVS-TL · COCO-TLV + ImageNet-TLV, A and B | F1: 0.85 -> 0.12 | Hurt |
| use temporal-logic reasoning instead of GPT-4 reasoning over per-frame annotations | NSVS-TL · Waymo and NuScenes TLV | F1: 9-15% F1 improvement | Improved |
| use TL reasoning instead of LLM reasoning for single-event queries | NSVS-TL · COCO-TLV/ImageNet-TLV Always/Eventually A | F1 | No clear effect |
| increase video length to 1000+ seconds for GPT-4 reasoning versus TL reasoning | GPT-4 baseline vs NSVS-TL · synthetic long videos, A until B | F1 | Improved |
Closest papers by meaning
- Before It Fades: Reinforcing Temporal Representations at Inference Time in VideoLLMs (NeurIPS 2026 · similarity 0.89)
- Logic-in-Frames: Dynamic Keyframe Search via Visual Semantic-Logical Verification for Long Video Understanding (NeurIPS 2025 · similarity 0.89)
- ReWind: Understanding Long Videos with Instructed Learnable Memory (CVPR 2025 · similarity 0.89)
- Enhancing Temporal Understanding in Video-LLMs through Stacked Temporal Attention in Vision Encoders (NeurIPS 2025 · similarity 0.89)
- Leveraging Textual Memory and Key Frame Reasoning for Full Video Understanding Using Off-the-Shelf LLMs and VLMs (Student Abstract) (AAAI 2025 · similarity 0.89)
- Re-thinking Temporal Search for Long-Form Video Understanding (CVPR 2025 · similarity 0.89)
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.
