APM-Bench: Benchmarking Cross-session Persistent Memory for Egocentric Streaming Video Assistants

arXiv:2609.37559 · cs.CV · Submitted 2026-09-29 · Read on arXiv

cs.CV

Submitted: 2026-09-29

Updated: 2026-09-29

Terminology

• Adaptive Response — 33× in this paper · explained in 1 episode
• Accuracy — 24× in this paper · explained in 1 episode
• Benchmark — 19× in this paper · explained in 2 episodes
• KV cache — 9× in this paper · explained in 3 episodes
• Benchmarking — 5× in this paper · explained in 3 episodes
• Conference on Computer Vision and Pattern Recognition — 5× in this paper
• Verification — 5× in this paper · explained in 2 episodes
• Grounding — 4× in this paper · explained in 3 episodes
• Fidelity — 3× in this paper · explained in 2 episodes
• Episodic Recall — 2× in this paper
• Adaptation — 1× in this paper · explained in 1 episode
• CVF International Conference on Computer Vision — 1× in this paper
• Core — 1× in this paper · explained in 2 episodes
• Entity State Tracking — 1× in this paper
• Evidence-Ready Answering — 1× in this paper
• Foundation Model — 1× in this paper · explained in 3 episodes
• Foundation Models — 1× in this paper · explained in 7 episodes
• Full-duplex interaction — 1× in this paper · explained in 2 episodes
• LLM-as-a-Judge — 1× in this paper · explained in 3 episodes
• Memory-Grounded Proactive Assistance — 1× in this paper
• Optical Character Recognition — 1× in this paper
• Reasoning Traces — 1× in this paper · explained in 3 episodes
• Registered-Condition Response — 1× in this paper
• Task Progress Guidance — 1× in this paper
• Temporal Reasoning — 1× in this paper · explained in 2 episodes
• Trace — 1× in this paper · explained in 2 episodes

Sources

• Realtime-Venus: A full-duplex interaction system with asynchronous delegation
• Qwen3-VL Technical Report
• Seed2.0 Model Card: Towards Intelligence Frontier for Real-World Complexity
• StreamTTT: Reconciling Real-Time Perception and Long-Term Memory in Streaming VLMs
• ObjectStream: Latent Objects as Memory Anchors for Streaming Video Understanding
• What Should a Streaming Video Model Remember?
• Video Streaming Thinking: VideoLLMs Can Watch and Think Simultaneously
• OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video Understanding?
• StreamingBench: Assessing the Gap for MLLMs to Achieve Streaming Video Understanding
• An Efficient Streaming Video Understanding Framework with Agentic Control
• PhoStream: Benchmarking Real-World Streaming for Omnimodal Assistants in Mobile Scenarios
• HD-EPIC: A Highly-Detailed Egocentric Video Dataset
• EgoPro-Bench: Benchmarking Personalized Proactive Interaction in Egocentric Video Streams
• A Simple Baseline for Streaming Video Understanding
• Vinci2: Providing Proactive Assistance in Continuous Egocentric Videos
• video-SALMONN S: Memory-Enhanced Streaming Audio-Visual LLM
• Think While Watching: Online Streaming Segment-Level Memory for Multi-Turn Video Reasoning in Multimodal Large Language Models
• InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency
• JoyAI-VL-Interaction: Real-Time Vision-Language Interaction Intelligence

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models