arXiv subjects

Wen Xia

Publications and source records attributed to Wen Xia.

10 recordsLinked to original sources

arXiv2026

High-accuracy Low-Bit KV-Cache Quantization via Local Distribution Restoration

Long-context large language model inference relies on the KV cache to avoid redundant attention computation, but incurs high memory and bandwidth overheads. Low-bit KV-cache quantization reduces this cost, yet it severely degrade quality; particularly, one-bit quantization reduces accuracy from 84.2% to 47.8% on Llama-3.1-8B under RULER. Rather than common beliefs that absolute error of logits, we find that the root cause is structured local misranking, where the distribution of logits in top-K region is drifted. We thereby propose local distribution restoration, a new technique that detects steps with high local distribution risk from quantized-logit features and restores only the selected top-K candidate distribution before token selection. We implement DGAP to achieve local distribution restoration, with efficient risk detcetors and correctors. Expeirments show that on Llama-3.1-8B, DGAP recovers K1V1 RULER accuracy from 47.8% to 83.2% and reduces distribution drift from 0.38 to 0.14; across Llama, Mistral, and Qwen models, it preserves the persistent low-bit KV-cache footprint with modest decode overhead.

cs.LG↗

arXiv2026

TTKV: Temporal-Tiered KV Cache for Long-Context LLM Inference

Key-value (KV) caching is critical for efficient inference in large language models (LLMs), yet its memory footprint scales linearly with context length, resulting in a severe scalability bottleneck. Existing approaches largely treat KV states as equally important across time, implicitly assuming uniform precision and accessibility. However, this assumption contrasts with human memory systems, where memories vary in clarity, recall frequency, and relevance with temporal proximity.Motivated by this insight, we propose TTKV, a KV cache management framework that maps the human memory system onto the KV cache. TTKV partitions the KV cache into temporal tiers with heterogeneous capacity and precision. The design addresses three aspects: (1) Tier Layout, decoupling fast and slow memory using HBM and DRAM; (2) Tier Content, assigning more recent KV states to faster, higher-precision tiers based on temporal proximity; and (3) Tier Interaction, employing block-wise streaming attention to overlap communication and computation when accessing slow tiers. Experiments show that TTKV reduces cross-tier traffic by 5.94x on 128K-context tasks, achieving up to 76% latency reduction and 2x throughput improvement over strong baselines.

cs.CL↗

arXiv2026

ProphetKV: User-Query-Driven Selective Recomputation for Efficient KV Cache Reuse in Retrieval-Augmented Generation

The prefill stage of long-context Retrieval-Augmented Generation (RAG) is severely bottlenecked by computational overhead. To mitigate this, recent methods assemble pre-calculated KV caches of retrieved RAG documents (by a user query) and reprocess selected tokens to recover cross-attention between these pre-calculated KV caches. However, we identify a fundamental "crowding-out effect" in current token selection criteria: globally salient but user-query-irrelevant tokens saturate the limited recomputation budget, displacing the tokens truly essential for answering the user query and degrading inference accuracy. We propose ProphetKV, a user-query-driven KV Cache reuse method for RAG scenarios. ProphetKV dynamically prioritizes tokens based on their semantic relevance to the user query and employs a dual-stage recomputation pipeline to fuse layer-wise attention metrics into a high-utility set. By ensuring the recomputation budget is dedicated to bridging the informational gap between retrieved context and the user query, ProphetKV achieves high-fidelity attention recovery with minimal overhead. Our extensive evaluation results show that ProphetKV retains 96%-101% of full-prefill accuracy with only a 20% recomputation ratio, while achieving accuracy improvements of 8.8%-24.9% on RULER and 18.6%-50.9% on LongBench over the state-of-the-art approaches (e.g., CacheBlend, EPIC, and KVShare).

cs.OS↗

arXiv2025

HybridEP: Scaling Expert Parallelism to Cross-Datacenter Scenario via Hybrid Expert/Data Transmission

Mixture-of-Experts (MoE) has become a popular architecture for scaling large models. However, the rapidly growing scale outpaces model training on a single DC, driving a shift toward a more flexible, cross-DC training paradigm. Under this, Expert Parallelism (EP) of MoE faces significant scalability issues due to the limited cross-DC bandwidth. Specifically, existing EP optimizations attempt to overlap data communication and computation, which has little benefit in low-bandwidth scenarios due to a much longer data communication time. Therefore, the trends of cross-DC EP scaling is fast becoming a critical roadblock to the continued growth of MoE models. To address this, we propose HybridEP, a modeling-guided framework to optimize EP under constrained bandwidth. Our key idea is to dynamically transform the spatial placement of experts to reduce data communication traffic and frequency, thereby minimizing EP's communication overheads. However, it is non-trivial to find the optimal solution because it complicates the original communication pattern by mixing data and expert communication. We therefore build a stream-based model to determine the optimal transmission ratio. Guided by this, we incorporate two techniques: (1) domain-based partition to construct the mapping between hybrid patterns and specific communication topology at GPU level, and (2) parameter-efficient migration to further refine this topology by reducing expert transmission overhead and enlarging the domain size. Combining all these designs, HybridEP can be considered as a more general EP with better scalability. Experimental results show that HybridEP outperforms existing state-of-the-art MoE training systems by up to 5.6x under constrained bandwidth. We further compare HybridEP and EP on large-scale simulations. HybridEP achieves up to 1.45x speedup with 1k DCs under different bandwidths.

cs.DC↗

arXiv2025

Critical $(P_5,W_4)$-Free Graphs

A graph $G$ is $k$-vertex-critical if $\chi(G) = k$ but $\chi(G-v)<k$ for all $v \in V(G)$. A graph is $(H_1,H_2)$-free if it contains no induced subgraph isomorphic to $H_1$ nor $H_2$. A $W_4$ is the graph consisting of a $C_4$ plus an additional vertex adjacent to all the vertices of the $C_4$. We show that there are finitely many $k$-vertex-critical $(P_5,W_4)$-free graphs for all $k \ge 1$ and we characterize all $5$-vertex-critical $(P_5,W_4)$-free graphs. Our results imply the existence of a polynomial-time certifying algorithm to decide the $k$-colorability of $(P_5,W_4)$-free graphs for each $k \ge 1$ where the certificate is either a $k$-coloring or a $(k+1)$-vertex-critical induced subgraph.

math.CO↗

arXiv2024

A Survey on Web Application Testing: Over a Decade of Evolution

As one of the most popular software applications, a web application is a program accessible through the web that dynamically generates content based on user interactions or contextual data; examples include online shopping platforms, social networking sites, and financial services. Web applications operate in diverse environments and leverage web technologies such as HTML, CSS, JavaScript, and Ajax, often incorporating features like asynchronous operations to enhance user experience. Due to the growing number of users and the popularity of web applications, the quality of these applications has become increasingly important. Web Application Testing (WAT) plays a vital role in ensuring the functionality, security, and reliability of web applications. Given the speed with which web technologies are evolving, WAT is especially important. In the last twelve years, various WAT approaches have been developed. The diversity of approaches reflects the many aspects of web applications, such as dynamic content, asynchronous operations, and diverse user environments. This paper provides a comprehensive overview of the main achievements over the last twelve years: It examines the main steps involved in WAT, including test case generation and execution, as well as evaluation and assessment. The currently available tools for WAT are also examined. The paper also discusses open research challenges and potential future work in WAT.

cs.SE↗

arXiv2024

Some Results on Critical ($P_5,H$)-free Graphs

Given two graphs $H_1$ and $H_2$, a graph is $(H_1,H_2)$-free if it contains no induced subgraph isomorphic to $H_1$ nor $H_2$. A graph $G$ is $k$-vertex-critical if every proper induced subgraph of $G$ has chromatic number less than $k$, but $G$ has chromatic number $k$. The study of $k$-vertex-critical graphs for specific graph classes is an important topic in algorithmic graph theory because if the number of such graphs that are in a given hereditary graph class is finite, then there exists a polynomial-time certifying algorithm to decide the $k$-colorability of a graph in the class. In this paper, we show that: (1) for $k \ge 1$, there are finitely many $k$-vertex-critical $(P_5,K_{1,4}+P_1)$-free graphs; (2) for $s \ge 1$, there are finitely many 5-vertex-critical $(P_5,K_{1,s}+P_1)$-free graphs; (3) for $k \ge 1$, there are finitely many $k$-vertex-critical $(P_5,\overline{K_3+2P_1})$-free graphs. Moreover, we characterize all $5$-vertex-critical $(P_5,H)$-free graphs where $H \in \{K_{1,3}+P_1,K_{1,4}+P_1,\overline{K_3+2P_1}\}$ using an exhaustive graph generation algorithm.

math.CO↗

arXiv2023

Critical $(P_5,dart)$-Free Graphs

Given two graphs $H_1$ and $H_2$, a graph is $(H_1,H_2)$-free if it contains no induced subgraph isomorphic to $H_1$ nor $H_2$. Let $P_t$ be the path on $t$ vertices. A dart is the graph obtained from a diamond by adding a new vertex and making it adjacent to exactly one vertex with degree 3 in the diamond. In this paper, we show that there are finitely many $k$-vertex-critical $(P_5,dart)$-free graphs for $k \ge 1$ To prove these results, we use induction on $k$ and perform a careful structural analysis via Strong Perfect Graph Theorem combined with the pigeonhole principle based on the properties of vertex-critical graphs. Moreover, for $k \in \{5, 6, 7\}$ we characterize all $k$-vertex-critical $(P_5,dart)$-free graphs using a computer generation algorithm. Our results imply the existence of a polynomial-time certifying algorithm to decide the $k$-colorability of $(P_5,dart)$-free graphs for $k \ge 1$ where the certificate is either a $k$-coloring or a $(k+1)$-vertex-critical induced subgraph.

math.CO↗

arXiv2022

Critical ($P_5$,bull)-free graphs

Given two graphs $H_1$ and $H_2$, a graph is $(H_1,H_2)$-free if it contains no induced subgraph isomorphic to $H_1$ or $H_2$. Let $P_t$ and $C_t$ be the path and the cycle on $t$ vertices, respectively. A bull is the graph obtained from a triangle with two disjoint pendant edges. In this paper, we show that there are finitely many 5-vertex-critical ($P_5$,bull)-free graphs.

math.CO↗

arXiv2017

Bandwidth-efficient Storage Services for Mitigating Side Channel Attack

Data deduplication is able to effectively identify and eliminate redundant data and only maintain a single copy of files and chunks. Hence, it is widely used in cloud storage systems to save storage space and network bandwidth. However, the occurrence of deduplication can be easily identified by monitoring and analyzing network traffic, which leads to the risk of user privacy leakage. The attacker can carry out a very dangerous side channel attack, i.e., learn-the-remaining-information (LRI) attack, to reveal users' privacy information by exploiting the side channel of network traffic in deduplication. Existing work addresses the LRI attack at the cost of the high bandwidth efficiency of deduplication. In order to address this problem, we propose a simple yet effective scheme, called randomized redundant chunk scheme (RRCS), to significantly mitigate the risk of the LRI attack while maintaining the high bandwidth efficiency of deduplication. The basic idea behind RRCS is to add randomized redundant chunks to mix up the real deduplication states of files used for the LRI attack, which effectively obfuscates the view of the attacker, who attempts to exploit the side channel of network traffic for the LRI attack. Our security analysis shows that RRCS could significantly mitigate the risk of the LRI attack. We implement the RRCS prototype and evaluate it by using three large-scale real-world datasets. Experimental results demonstrate the efficiency and efficacy of RRCS.

cs.CR↗