arXiv subjects

Zhifang Li

Publications and source records attributed to Zhifang Li.

6 recordsLinked to original sources

arXiv2026

SieveIVF: Threshold-Aware IVF Execution for Large-Scale Training Data Deduplication

Embedding-based training data deduplication retrieves candidate duplicate edges above an application similarity threshold, but fixed-probe inverted-file (IVF) search ignores this predicate when giving every query the same partition budget. Across four Hunyuan workloads, qualifying neighbors appear early despite sharply varying search depths. We present SieveIVF, a threshold-aware IVF executor that stops after $W$ consecutive searches find no qualifying candidate. The systems challenge is to preserve partition-major batching when each query's remaining work depends on prior results. Continuous batching groups ready queries by partition. A lookahead scheduler layers on top, exposing only work committed by the stopping rule to increase concurrency without changing stopping decisions or returned results. We implement SieveIVF in Lance. At $W=8$, SieveIVF is $4.1$--$7.6\times$ faster than fixed-probe IVF on four 10M Hunyuan workloads and $6.1$--$8.4\times$ faster on two public 100M workloads under the same index and search parameters, with pooled filtered top-10 recall losses of $0.03$--$1.13$ percentage points on Hunyuan and $1.43$--$2.29$ percentage points on the public workloads. These results show how an application predicate can guide IVF work allocation without changing the index or bounded top-$k$ interface.

cs.DB↗

arXiv2022

Game-theoretical approach for opinion dynamics on social networks

Opinion dynamics on social networks have been received considerable attentions in recent years. Nevertheless, just a few works have theoretically analyzed the condition in which a certain opinion can spread in the whole structured population. In this paper, we propose an evolutionary game approach for a binary opinion model to explore the conditions for an opinion's spreading. Inspired by real-life observations, we assume that an agent's choice to select an opinion is not random, but is based on a score rooted both from public knowledge and the interactions with neighbors. By means of coalescing random walks, we obtain a condition in which opinion $A$ can be favored to spread on social networks in the weak selection limit. We find that the successfully spreading condition of opinion $A$ is closely related to the basic scores of binary opinions, the feedback scores on opinion interactions, and the structural parameters including the edge weights, the weighted degrees of vertices, and the average degree of the network. In particular, when individuals adjust their opinions based solely on the public information, the vitality of opinion $A$ depends exclusively on the difference of basic scores of $A$ and $B$. When there are no negative (positive) feedback interactions between connected individuals, we find that the success of opinion $A$ depends on the ratio of the obtained positive (negative) feedback scores of competing opinions. To complete our study, we perform computer simulations on fully-connected, small-world, and scale-free networks, respectively, which support and confirm our theoretical findings.

cs.GT↗

arXiv2017

A Straightforward Method to Judge the Completeness of a Polymorphic Gate Set

Polymorphic circuits are a special kind of circuits which possess some different build-in functions and these functions are activated by environment parameters, like light and VDD. Some theories have been proposed to guide the design of polymorphic circuits, including the definition of complete polymorphic gate sets and algorithms to judge the completeness of a polymorphic gate set. However, the previous algorithms have to enumerate all the polymorphic signals for judging the completeness of a polymorphic gate set, and it is not easy to be conducted manually. In this paper, a straightforward method is proposed to judge the completeness of a polymorphic gate set. And the correctness of the straightforward method is proved theoretically. Some examples are given to show that the proposed method could be conducted step by step. Its actual computing cost is usually low, and it is suitable for manual operation.

cs.ET↗

arXiv2017

A Simplification Method of Polymorphic Boolean Functions

Polymorphic circuits are a special kind of circuits which possess multiple build-in functions, and these functions are activated by environment parameters, like temperature, light and VDD. The behavior of a polymorphic circuit can be described by a polymorphic Boolean function. For the first time, this brief presents a simplification method of the polymorphic Boolean function.

cs.ET↗

arXiv2017

Design Methods for Polymorphic Combinational Logic Circuits based on the Bi_Decomposition Approach

Polymorphic circuits are a special kind of digital logic components, which possess multiple build-in functions. In different environments, a polymorphic circuit would perform different functions. Evolutionary Algorithms, Binary Decision Diagrams (BDD) and the multiplex method have been adopted to design polymorphic circuits. However, the evolutionary methods face the scalable problem. The BDD method consumes too much gate resource. The polymorphic circuit built by the multiplex method rarely contains polymorphic gates. In this paper, based on the traditional Bi_Decomposition circuit design approach, two methods, i.e. the Poly_Bi_Decomposition method and the Transformation&Bi_Decomposition method, are proposed for designing polymorphic circuits. The Poly_Bi_Decomposition method can design relatively large and gate-efficient polymorphic circuits with a higher percentage of polymorphic gates. The Transformation&Bi_Decomposition method can use the traditional circuit design approaches and tools, e.g. Bi_Decomposition, to design polymorphic circuits directly. The experimental results show the good performance of the proposed methods.

cs.ET↗

arXiv2011

Detecting the optimal number of communities in complex networks

To obtain the optimal number of communities is an important problem in detecting community structure. In this paper, we extend the measurement of community detecting algorithms to find the optimal community number. Based on the normalized mutual information index, which has been used as a measure for similarity of communities, a statistic $Ω(c)$ is proposed to detect the optimal number of communities. In general, when $Ω(c)$ reaches its local maximum, especially the first one, the corresponding number of communities \emph{c} is likely to be optimal in community detection. Moreover, the statistic $Ω(c)$ can also measure the significance of community structures in complex networks, which has been paid more attention recently. Numerical and empirical results show that the index $Ω(c)$ is effective in both artificial and real world networks.

physics.soc-ph↗