arXiv subjects

Zhenyu Ning

Publications and source records attributed to Zhenyu Ning.

5 recordsLinked to original sources

arXiv2026

LiveVLM: Efficient Online Video Understanding via Streaming-Oriented KV Cache and Retrieval

Recent developments in Video Large Language Models (Video LLMs) have enabled models to process hour-long videos and exhibit exceptional performance. Nonetheless, the Key-Value (KV) cache expands linearly over time, leading to substantial memory overhead and response delay--critical challenges in various real-world online applications, such as Deepseek services, autonomous driving and robotics. To mitigate these issues, we propose $\textbf{LiveVLM}$, a training-free and query-agnostic framework specifically designed for online video understanding and real-time interaction. LiveVLM employs a Vision Sink Bucketing (VSB) mechanism to process video streams in real time, retain long-term video details and eliminate redundant KVs. This mechanism utilizes vision-to-vision attention scores as the metric and seeks to maximize the coverage of contextual information during compression. Noting that KV cache compressed in a query-agnostic manner inevitably retains irrelevant information for specific queries, LiveVLM incorporates a Position-agnostic KV Retrieval (PaR) mechanism to reduce interference from redundant context. The keypoint of PaR lies in decoupling positional embeddings to enhance the similarity between key tensors, thereby supporting efficient retrieval at the granularity of pages. Extensive experiments demonstrate that LiveVLM enables the foundation LLaVA-OneVision model to achieve state-of-the-art accuracy among both training-free query-agnostic methods and training-based online models.

cs.CV↗

arXiv2026

FreeKV: Boosting KV Cache Retrieval for Efficient LLM Inference

Large language models (LLMs) are widely deployed with rapidly expanding context windows to support increasingly demanding applications. However, long contexts pose significant deployment challenges, primarily due to the KV cache whose size grows proportionally with context length. While KV cache compression methods have been proposed to address this issue, KV dropping methods incur considerable accuracy loss, and KV retrieval methods suffer from significant efficiency bottlenecks. We propose FreeKV, a training-free algorithm-system co-optimization framework to enhance KV retrieval efficiency while preserving accuracy. On the algorithm side, FreeKV introduces speculative retrieval to shift the KV selection and recall processes out of the critical path, combined with fine-grained correction to ensure accuracy. On the system side, FreeKV employs hybrid KV layouts across CPU and GPU memory to eliminate fragmented data transfers, and leverages double-buffered streamed recall to further improve efficiency, enabling effective overlap with computation, full latency hiding, and practical speedups from speculative recall. Experiments demonstrate that FreeKV achieves near-lossless accuracy across various scenarios and models, delivering up to a 13$\times$ speedup compared to SOTA KV retrieval methods. Code is available at https://github.com/sjtu-zhao-lab/FreeKV.

cs.LG↗

arXiv2024

Inf-MLLM: Efficient Streaming Inference of Multimodal Large Language Models on a Single GPU

Multimodal Large Language Models (MLLMs) are distinguished by their multimodal comprehensive ability and widely used in many real-world applications including GPT-4o, autonomous driving and robotics. Despite their impressive performance, the multimodal inputs always incur long context. The inference under long context requires caching massive Key and Value states (KV cache) of previous tokens, which introduces high latency and excessive memory consumption. Due to this reason, it is challenging to deploy streaming inference of MLLMs on edge devices, which largely constrains the power and usage of MLLMs in real-world applications. In this paper, we introduce Inf-MLLM, an efficient inference framework for MLLMs, which enable streaming inference of MLLM on a single GPU with infinite context. Inf-MLLM is based on our key observation of the attention pattern in both LLMs and MLLMs called "attention saddles". Thanks to the newly discovered attention pattern, Inf-MLLM maintains a size-constrained KV cache by dynamically caching recent tokens and relevant tokens. Furthermore, Inf-MLLM proposes attention bias, a novel approach to enable MLLMs to capture long-term dependency. We show that Inf-MLLM enables multiple LLMs and MLLMs to achieve stable performance over 4M-token long texts and multi-round conversations with 1-hour-long videos on a single GPU. In addition, Inf-MLLM exhibits superior streaming reasoning quality than existing methods such as StreamingLLM and 2x speedup than H2O.

cs.LG↗

arXiv2024

The CAP Principle for LLM Serving: A Survey of Long-Context Large Language Model Serving

We survey the large language model (LLM) serving area to understand the intricate dynamics between cost-efficiency and accuracy, which is magnified by the growing need for longer contextual understanding when deploying models at a massive scale. Our findings reveal that works in this space optimize along three distinct but conflicting goals: improving serving context length (C), improving serving accuracy (A), and improving serving performance (P). Drawing inspiration from the CAP theorem in databases, we propose a CAP principle for LLM serving, which suggests that any optimization can improve at most two of these three goals simultaneously. Our survey categorizes existing works within this framework. We find the definition and continuity of user-perceived measurement metrics are crucial in determining whether a goal has been met, akin to prior CAP databases in the wild. We recognize the CAP principle for LLM serving as a guiding principle, rather than a formal theorem, to inform designers of the inherent and dynamic trade-offs in serving models. As serving accuracy and performance have been extensively studied, this survey focuses on works that extend serving context length and address the resulting challenges.

cs.DB↗

arXiv2018

DexLego: Reassembleable Bytecode Extraction for Aiding Static Analysis

The scale of Android applications in the market is growing rapidly. To efficiently detect the malicious behavior in these applications, an array of static analysis tools are proposed. However, static analysis tools suffer from code hiding techniques like packing, dynamic loading, self modifying, and reflection. In this paper, we thus present DexLego, a novel system that performs a reassembleable bytecode extraction for aiding static analysis tools to reveal the malicious behavior of Android applications. DexLego leverages just-in-time collection to extract data and bytecode from an application at runtime, and reassembles them to a new Dalvik Executable (DEX) file offline. The experiments on DroidBench and real-world applications show that DexLego correctly reconstructs the behavior of an application in the reassembled DEX file, and significantly improves analysis result of the existing static analysis systems.

cs.CR↗