REVIEW 25 cited by
Streaming Video Question-Answering with In-context Video KV-Cache Retrieval
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Streaming Video Question-Answering with In-context Video KV-Cache Retrieval
read the original abstract
We propose ReKV, a novel training-free approach that enables efficient streaming video question-answering (StreamingVQA), by seamlessly integrating with existing Video Large Language Models (Video-LLMs). Traditional VideoQA systems struggle with long videos, as they must process entire videos before responding to queries, and repeat this process for each new question. In contrast, our approach analyzes long videos in a streaming manner, allowing for prompt responses as soon as user queries are received. Building on a common Video-LLM, we first incorporate a sliding-window attention mechanism, ensuring that input frames attend to a limited number of preceding frames, thereby reducing computational overhead. To prevent information loss, we store processed video key-value caches (KV-Caches) in RAM and disk, reloading them into GPU memory as needed. Additionally, we introduce a retrieval method that leverages an external retriever or the parameters within Video-LLMs to retrieve only query-relevant KV-Caches, ensuring both efficiency and accuracy in question answering. ReKV enables the separation of video encoding and question-answering across different processes and GPUs, significantly enhancing the efficiency of StreamingVQA. Through comprehensive experimentation, we validate the efficacy and practicality of our approach, which significantly boosts efficiency and enhances applicability over existing VideoQA models.
Forward citations
Cited by 25 Pith papers
-
EGOSTREAM: A Diagnostic Benchmark for Streaming Episodic Memory in Egocentric Vision
Egostream introduces a diagnostic benchmark that expands 2,250 questions into 8,528 recall-conditioned evaluations to measure streaming episodic memory performance across detail, spatial, temporal, event, social, caus...
-
Don't Pause! Every prediction matters in a streaming video
SPOT-Bench tests real-time streaming video perception with timeliness metrics, exposing limitations in current models and introducing AsynKV as an improved baseline.
-
Mosaic: Cross-Modal Clustering for Efficient Video Understanding
Mosaic uses cross-modal clusters as the unit for KVCache organization in VLMs to achieve up to 1.38x speedup in streaming long-video understanding.
-
STAC: Plug-and-Play Spatio-Temporal Aware Cache Compression for Streaming 3D Reconstruction
STAC compresses KV caches in streaming 3D reconstruction transformers via temporal token preservation with decayed attention, spatial voxel compression, and chunked multi-frame optimization, delivering 10x memory redu...
-
Can Multi-Modal LLMs Provide Live Step-by-Step Task Guidance?
Introduces the first dedicated benchmark for live multi-modal LLM task guidance with mistake detection and a streaming baseline model.
-
SFHand: Learning Embodied Manipulation by Streaming Egocentric 3D Hand Forecasting
SFHand presents the first streaming language-guided autoregressive framework for 3D hand forecasting, achieving up to 35.8% gains over prior methods and 13.4% better downstream embodied task performance.
-
ReToken: One Token to Improve Vision-Language Models for Visual Retrieval
A single learnable token scored against final-layer value vectors retrieves query-relevant visual KV cache entries and lifts VLM accuracy on image haystacks and hour-long video.
-
ObjectStream: Latent Objects as Memory Anchors for Streaming Video Understanding
A training-free memory framework that anchors streaming video memory to latent objects discovered from frozen Video-LLM features, improving streaming QA accuracy while cutting memory and latency.
-
ObjectStream: Latent Objects as Memory Anchors for Streaming Video Understanding
Training-free latent-object memory anchors let frozen Video-LLMs retain object histories under a tight token budget and improve streaming and long-video QA.
-
ChronoStitch: Training-Free Composition of Visual KV Memories for Long-Horizon Temporal Reasoning
ChronoStitch re-bases stored video-chunk KV keys into a global multimodal RoPE frame and selectively recomputes a small slice of high-deviation tokens, recovering most of the joint-prefill temporal-reasoning gap at 3....
-
ReflectWorld-MM: An Entity-Oriented Multimodal Memory System for Open-Ended Video Streams
An entity-oriented hierarchical multi-media memory system for open-ended video streams achieves best accuracy on six long-video and lifelong-memory benchmarks.
-
StreamOV: Streaming Omni-Video Understanding via Evidence-Guided Memory and Response Triggering
StreamOV proposes evidence-guided long-short term memory and a hidden-state-driven trigger for efficient online audio-visual reasoning in streaming videos, along with the SOVBench benchmark for multi-turn evaluation.
-
Response-G1: Explicit Scene Graph Modeling for Proactive Streaming Video Understanding
Response-G1 uses query-guided scene graphs, memory retrieval, and augmented prompting to improve when Video-LLMs decide to respond during streaming videos.
-
HERMES: KV Cache as Hierarchical Memory for Efficient Streaming Video Understanding
HERMES organizes the KV cache into a hierarchical memory to enable real-time streaming video understanding in MLLMs, achieving 10x faster TTFT and up to 11.4% accuracy gains on streaming benchmarks with 68% fewer tokens.
-
LinMU: Multimodal Understanding Made Linear
LinMU achieves linear-complexity multimodal understanding by swapping self-attention for an M-MATE dual-branch block and distilling from a frozen teacher VLM, matching accuracy with up to 2.7x faster TTFT and 9x highe...
-
OmniZip: Audio-Guided Dynamic Token Compression for Fast Omnimodal Large Language Models
OmniZip introduces an audio-guided dynamic token compression framework that achieves 3.42X inference speedup and 1.4X memory reduction for omnimodal LLMs without any training.
-
StreamingVLM: Real-Time Understanding for Infinite Video Streams
StreamingVLM enables stable real-time understanding of infinite video streams at up to 8 FPS using a streaming KV cache and aligned SFT on overlapped chunks, with a 66.18% win rate over GPT-4O mini on a new two-hour v...
-
LiveVLM: Efficient Online Video Understanding via Streaming-Oriented KV Cache and Retrieval
LiveVLM introduces VSB and PaR to compress and retrieve KV cache in streaming video LLMs, enabling LLaVA-OneVision to reach SOTA accuracy among training-free query-agnostic and training-based online models.
-
ReflectWorld-MM: An Entity-Oriented Multimodal Memory System for Open-Ended Video Streams
An entity-oriented hierarchical multimodal memory system for open-ended video streams reports best accuracy on six long-video and lifelong-memory benchmarks.
-
OmniFocus: Query-Guided Modality-Balanced Token Compression for Omni-Modal Large Language Models
Query-guided, modality-symmetric token compression preserves audio-visual evidence better than audio-guided budgets at 25–35% retention on Qwen2.5-Omni.
-
OmniMem: Perturbation-aware Memory Compression for Streaming Audio-Visual LLMs
OmniMem achieves 2-4% higher accuracy than training-free baselines on long video benchmarks for audio-visual LLMs by using modality-aware KV cache allocation and perturbation-aware state selection, with further gains ...
-
Response-G1: Explicit Scene Graph Modeling for Proactive Streaming Video Understanding
Response-G1 uses query-guided scene graph generation, memory retrieval, and retrieval-augmented prompting to improve proactive response timing in streaming video understanding.
-
VLMaxxing through FrameMogging Training-Free Anti-Recomputation for Video Vision-Language Models
Training-free adaptive reuse of stable visual state in video VLMs reduces follow-up latency by 15-36x on Qwen2.5-VL while preserving correctness on VideoMME, with smaller first-query speedups via pruning.
-
Towards a Dynamic and Fixed-budget Memory Bank for Efficient Streaming Video Understanding
CausalMem constructs a dynamic fixed-budget memory bank for streaming video in MLLMs via online semantic basis updates, achieving 20x token compression and accuracy gains on benchmarks when applied to LLaVA-OneVision ...
-
Exploring Multimodal LMMs for Online Episodic Memory Question Answering on the Edge
An edge-deployed multimodal LLM pipeline for online episodic memory QA reaches 51.76% accuracy on an 8 GB GPU and 54.40% on a local server, within 4-5 points of a 56% cloud baseline.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.