# Semantic-Aware Adaptive Visual Memory for Streaming Video Understanding

arXiv: https://arxiv.org/abs/2605.07897 · HTML: https://mlsanity.com/paper/2605.07897/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 45 of its 48 references that are in the map.
- First posted on arXiv 8 May 2026.

## Builds on (most cited first, 30 of 45)

- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding](https://mlsanity.com/paper/2306.02858/) (2023)
- [ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT](https://mlsanity.com/paper/2004.12832/) (2020)
- [Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models](https://mlsanity.com/paper/2306.05424/) (2023)
- [How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suites](https://mlsanity.com/paper/2404.16821/) (2024)
- [Token Merging: Your ViT But Faster](https://mlsanity.com/paper/2210.09461/) (2022)
- [LongVideoBench: A Benchmark for Long-context Interleaved Video-Language Understanding](https://mlsanity.com/paper/2407.15754/) (2024)
- [LLaVA-Video: Video Instruction Tuning With Synthetic Data](https://mlsanity.com/paper/2410.02713/) (2024)
- [Long Context Transfer from Language to Vision](https://mlsanity.com/paper/2406.16852/) (2024)
- [End-to-end Autonomous Driving: Challenges and Frontiers](https://mlsanity.com/paper/2306.16927/) (2023)
- [LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding](https://mlsanity.com/paper/2410.17434/) (2024)
- [VideoLLM-online: Online Video Large Language Model for Streaming Video](https://mlsanity.com/paper/2406.11816/) (2024)
- [InternVideo2.5: Empowering Video MLLMs with Long and Rich Context Modeling](https://mlsanity.com/paper/2501.12386/) (2025)
- [Video Understanding with Large Language Models: A Survey](https://mlsanity.com/paper/2312.17432/) (2023)
- [Aligning Cyber Space with Physical World: A Comprehensive Survey on Embodied AI](https://mlsanity.com/paper/2407.06886/) (2024)
- [Flash-VStream: Memory-Based Real-Time Understanding for Long Video Streams](https://mlsanity.com/paper/2406.08085/) (2024)
- [Streaming Long Video Understanding with Large Language Models](https://mlsanity.com/paper/2405.16009/) (2024)
- [To See is to Believe: Prompting GPT-4V for Better Visual Instruction Tuning](https://mlsanity.com/paper/2311.07574/) (2023)
- [StreamingBench: Assessing the Gap for MLLMs to Achieve Streaming Video Understanding](https://mlsanity.com/paper/2411.03628/) (2024)
- [OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video Understanding?](https://mlsanity.com/paper/2501.05510/) (2025)
- [Dispider: Enabling Video LLMs with Active Real-Time Interaction via Disentangled Perception, Decision, and Reaction](https://mlsanity.com/paper/2501.03218/) (2025)
- [A Survey of Context Engineering for Large Language Models](https://mlsanity.com/paper/2507.13334/) (2025)
- [Streaming Video Question-Answering with In-context Video KV-Cache Retrieval](https://mlsanity.com/paper/2503.00540/) (2025)
- [TimeChat-Online: 80% Visual Tokens are Naturally Redundant in Streaming Videos](https://mlsanity.com/paper/2504.17343/) (2025)
- [StreamingVLM: Real-Time Understanding for Infinite Video Streams](https://mlsanity.com/paper/2510.09608/) (2025)
- [StreamForest: Efficient Online Video Understanding with Persistent Event Memory](https://mlsanity.com/paper/2509.24871/) (2025)
- [ChatVideo: A Tracklet-centric Multimodal and Versatile Video Understanding System](https://mlsanity.com/paper/2304.14407/) (2023)

## Built on by (most cited first, 8 of 8)

- [ObjectStream: Latent Objects as Memory Anchors for Streaming Video Understanding](https://mlsanity.com/paper/2607.28312/) (2026)
- [Realtime-Venus: A full-duplex interaction system with asynchronous delegation](https://mlsanity.com/paper/2609.13814/) (2026)
- [ShallowStream: Index Shallow then Answer Deep for Streaming Video Understanding](https://mlsanity.com/paper/2609.02780/) (2026)
- [Beyond Retrieval: Progressive Latent Memory Evolution for Streaming Video Understanding](https://mlsanity.com/paper/2609.04131/) (2026)
- [MemEvo: Automatic Discovery of Streaming Video Memory Mechanisms](https://mlsanity.com/paper/2609.36581/) (2026)
- [When to Retrieve, When to Stay: Uncertainty-Aware Temporal Evidence Allocation for Streaming Video-LLMs](https://mlsanity.com/paper/2609.37345/) (2026)
- [APM-Bench: Benchmarking Cross-session Persistent Memory for Egocentric Streaming Video Assistants](https://mlsanity.com/paper/2609.37559/) (2026)
- [FlashBack: Knowing When to Remember in Streaming Vision-Language Models](https://mlsanity.com/paper/2610.01192/) (2026)

## Abstract (arXiv)

Online streaming video understanding requires models to process continuous visual inputs and respond to user queries in real time, where the unbounded stream and unpredictable query timing turn memory management into a central challenge. Existing methods typically compress visual tokens via visual similarity heuristics, or augment compression with KV-cache-level retrieval. However, compression decisions rarely incorporate semantic signals, and retrieval is often added after compression is finalized, making the two stages hard to coordinate. We present SAVEMem, a training-free dual-stage framework that brings semantic awareness into memory generation and lets the retrieval scope adapt per query. In Stage~1, SAVEMem builds a three-tier streaming memory online under a constant memory budget. A fixed pseudo-question bank provides a lightweight semantic prior, so that long-term retention is shaped by semantic salience rather than visual similarity alone. In Stage~2, SAVEMem performs query-aware retrieval over this memory. An anchor-conditioned recency gate adapts the retrieval scope from short-term to mid- and long-term memory based on whether the query targets the present or the distant past. Within this scope, late interaction between query and memory tokens selects candidate frames for answering. Applied to Qwen2.5-VL without training, SAVEMem improves the OVO-Bench overall score from 52.27 to 62.69 and yields consistent gains on StreamingBench and ODV-Bench, while reducing peak GPU memory by 48% at 128 frames over the backbone.