KV Cache

Series guide. This article is the bounded-memory thematic survey. Start with the development-history hub for the six-path map, read the ReKV note for the exact-archive baseline, and use the pipeline survey for the stage-by-stage architecture. A streaming Video-LMM has to make decisions before it knows which detail a future user will care about. ReKV avoids write-time eviction or compression by storing every generated visual KV block, then retrieves a small subset after the question is known. This does not recover unsampled pixels or information absent from those model states. Fixed-state methods make the opposite trade: they evict, merge, or summarize history during the stream so that memory no longer grows with video duration. ...

Paper: WeaveTime: Streaming from Earlier Frames into Emergent Memory in VideoLLMs Preprint (earlier title: Stream from Earlier Frames): arXiv:2602.22142 PDF: CVPR 2026 Open Access Supplement: CVPR 2026 Supplementary Material Project: WeaveTime Code: zhangyl4/weavetime Series guide. PCDF bounds when and how much history is read, not how much ReKV-style history is stored. See the bounded-memory survey for the resource comparison and the development-history hub for the complete series. Core Idea WeaveTime can be summarized as: ReKV-style historical visual KV memory + temporal-order SFT + uncertainty-gated coarse-to-fine retrieval. ...

Paper: StreamingVLM: Real-Time Understanding for Infinite Video Streams Project: StreamingVLM Code: mit-han-lab/streaming-vlm Datasets: Inf-Stream-Train / Inf-Stream-Eval Core Idea StreamingVLM is designed for a different target from most streaming VideoQA papers: it keeps watching a video and producing synchronized commentary, instead of waiting for a future question and then retrieving historical evidence. Its main idea is to align a simple streaming inference cache with the context pattern used during supervised fine-tuning. At inference time, the model keeps only: ...

Paper: MuKV: Multi-Grained KV Cache Compression for Long Streaming Video Question-Answering Proceedings: CVPR 2026 Open Access Code: IMBALDY/MuKV Series guide. A fixed per-question read can still sit on a growing multi-grained archive. See the bounded-memory survey for that accounting and the development-history hub for the complete series. Background Long streaming VideoQA has a simple but painful constraint: the video keeps arriving, while the future user questions are unknown. KV-cache methods such as ReKV make this setting more practical. Instead of recomputing historical video tokens when a question arrives, the model can prefill the video stream in advance, store the visual KV cache, retrieve the relevant cache blocks later, and answer with much lower online cost. ...

Series guide. This article is the stage-by-stage architecture map. Start with the development-history hub for the six research paths, and use the benchmark index for evaluation settings and protocols. Related papers: ReKV: Paper / Code StreamKV: Paper / Code InfiniPot-V: Paper / Code StreamMem: Paper LiveVLM: Paper / Code StreamingTOM: Paper / Code rLiVS: Paper / Code Core Question All of these papers are trying to solve the same systems problem: When a video stream keeps arriving, the user question is not known yet, and GPU memory is limited, how should a Video-LLM process the stream, compress memory, retrieve evidence, and generate an answer with low latency? ...

Paper: StreamingTOM: Streaming Token Compression for Efficient Video Understanding Code: YIGE24/StreamingTOM Background Streaming video understanding has two constraints that offline video understanding does not really need to respect: causality: the model cannot use future frames to decide how to compress current frames; accumulation: tokens and KV cache keep growing as the video stream becomes longer. Most recent training-free streaming methods mainly work on the post-LLM KV cache: ReKV stores historical KV blocks and retrieves relevant ones at question time; StreamKV improves the segmentation / compression / retrieval pipeline; InfiniPot-V and StreamMem keep a bounded KV memory with query-agnostic compression; LiveVLM combines query-agnostic KV compression with query-time retrieval. These methods are useful, but they still have one important blind spot: ...

Paper: LiveVLM: Efficient Online Video Understanding via Streaming-Oriented KV Cache and Retrieval Code: sjtu-zhao-lab/LiveVLM Acceptance: DAC 2026 program Version note. This article follows the substantially revised arXiv v2 (2026-04-23) / DAC 2026 method: Vision Sink Bucketing and Position-agnostic KV Retrieval. The 2025 v1 used a different chunk-compression and FIFO mechanism; its results should not be mixed with the v2 design. Series guide. For the distinction between bounded VSB state and query-time PaR reads, see the bounded-memory survey. The development-history hub places LiveVLM in the complete series. ...

Paper: StreamMem: Query-Agnostic KV Cache Memory for Streaming Video Understanding Project: StreamMem Series guide. For how proxy-guided retention fits between exact archives and summary-state memory, see the bounded-memory survey. The development-history hub places this method in the complete series. Background Streaming video understanding is hard because the model has to process frames as they arrive, without knowing: how long the video will be; what future user questions will ask; which past details will become important later. For long videos, the visual tokens and their KV cache keep growing over time. Even if a long-context MLLM can technically accept many tokens, storing and attending to all historical KV entries is still expensive. ...

Paper: InfiniPot-V: Memory-Constrained KV Cache Compression for Streaming Video Understanding Preprint: arXiv:2506.15745 Research reimplementation: aiha-lab/InfiniPot-V Series guide. For why a fixed retained KV cache differs from a bounded read over a growing archive, see the bounded-memory survey. The development-history hub places this shift in the complete series. Background Streaming video understanding is more constrained than offline long-video understanding. In offline settings, the model can see the whole video first, maybe even the user query first, and then decide how to compress tokens or KV cache. But in streaming settings: ...

Paper: StreamKV: Streaming Video Question-Answering with Segment-based KV Cache Retrieval and Compression Code: sou1p0wer/StreamKV Background Streaming video question-answering (StreamingVQA) requires a model to continuously process incoming video, preserve useful historical context, and answer questions online with low latency. ReKV showed that video QA can be reformulated as retrieve relevant KV caches first, then answer with the retrieved KV. But it still has several weaknesses: It uses uniform segmentation, which may cut through semantic boundaries. It keeps essentially the whole historical visual context, so memory usage is still large. Its retrieval strategy is not flexible enough, especially when the useful information is distributed differently across layers. Core Idea StreamKV extends the ReKV line in two directions at the same time: ...