Research map / Video / Video-language understanding
Keep It in Mind: User Centric Continual Spatial Intelligence Reasoning in Egocentric Video Streams
Yun wang, Junbin Xiao, Han Lyu, Yifan Wang, Jing Zuo, Zhanjie Zhang, Hong Huang, Dapeng Wu, Angela Yao
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 per-frame MLLM spatial descriptions as extra context (ablation) | Qwen3-VL-8B · UCS-Bench | overall accuracy: 44.0 -> 44.5 | No clear effect |
| add pose-anchored scene graph memory (ablation) | Qwen3-VL-8B · UCS-Bench | overall accuracy: 44.0 -> 50.2 | Improved |
| add spatial descriptions on top of scene graph (ablation) | DirectMe · UCS-Bench | overall accuracy: 50.2 -> 50.5 | No clear effect |
| wrap InternVL3-8B with DirectMe | InternVL3-8B · UCS-Bench | overall accuracy: 43.8 -> 49.7 | Improved |
| wrap Qwen3-VL-8B with DirectMe | Qwen3-VL-8B · UCS-Bench | overall accuracy: 44.0 -> 50.5 | Improved |
| use spatial-centric models instead of general MLLMs | Qwen3-VL-8B baseline · UCS-Bench | overall accuracy: SenseNova-SI 37.8, Spatial-MLLM 34.3 vs Qwen3-VL-8B 44.0 | Hurt |
| use streaming-specific architectures instead of general MLLMs | video QA baselines · UCS-Bench | overall accuracy: Flash-VStream 28.5, Dispider 36.7, VideoChat-Online 37.5 vs InternVL3-8B 43.8 | Hurt |
| scale InternVL3 from 8B to 38B | InternVL3 · UCS-Bench | overall accuracy: 43.8 -> 49.2 | Improved |
| scale Qwen3-VL from 8B to 32B | Qwen3-VL · UCS-Bench | overall accuracy: 44.0 -> 49.0 | Improved |
| scale InternVL3.5 from 8B to 38B | InternVL3.5 · UCS-Bench | overall accuracy: 42.9 -> 48.1 | Improved |
| increase evidence-to-question time gap | all models including DirectMe · UCS-Bench | accuracy | Hurt |
Closest papers by meaning
- EgoThinker: Unveiling Egocentric Reasoning with Spatio-Temporal CoT (NeurIPS 2025 · similarity 0.89)
- SAW-Bench: Learning Situated Awareness in the Real World (ICML 2026 · similarity 0.88)
- SAVVY: Spatial Awareness via Audio-Visual LLMs through Seeing and Hearing (NeurIPS 2025 · similarity 0.88)
- Seeing Together: Multi-Robot Cooperative Egocentric Spatial Reasoning with Multimodal Large Language Models (NeurIPS 2026 · similarity 0.87)
- Omnia de EgoTempo: Benchmarking Temporal Understanding of Multi-Modal LLMs in Egocentric Videos (CVPR 2025 · similarity 0.87)
- Learning to Evolve Scenes: Reasoning about Human Activities with Scene Graphs (NeurIPS 2026 · 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.
