Have I Scene This Before? Spatially Grounded Conversational Memory for Complex Queries in Egocentric Assistants
cs.AI
Submitted: 2026-10-04
Updated: 2026-10-04
Code: https://github.com/joeliang0520/SAPCR-MEM
Project page: https://video-mme.github.io
Sources
- Qwen3-VL Technical Report
- ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes
- M2A: Multimodal Memory Agent with Dual-Layer Hybrid Memory for Long-Term Personalized Interactions
- ConceptGraphs: Open-Vocabulary 3D Scene Graphs for Perception and Planning
- Towards Spatial Supersensing in the Wild
- Does Memory Need Graphs? A Unified Framework and Empirical Analysis for Long-Term Dialog Memory
- Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
- Goal-Oriented Reasoning for RAG-based Memory in Conversational Agentic LLM Systems
- Depth Anything 3: Recovering the Visual Space from Any Views
- Temporal Order Matters for Agentic Memory: Segment Trees for Long-Horizon Agents
- SegVGGT: Joint 3D Reconstruction and Instance Segmentation from Multi-View Images
- Adaptive Keyframe Sampling for Long Video Understanding
- LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory
- LongVideoBench: A Benchmark for Long-context Interleaved Video-Language Understanding
- Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems