TY - GEN

T1 - PySymGym: An Infrastructure to Train AI-Powered Navigation Assistant for Symbolic Execution Engine

AU - Григорьев, Семен Вячеславович

AU - Чистякова, Анна Артуровна

AU - Шеметова, Екатерина Николаевна

AU - Нигматулин, Максим Владиславович

AU - Парфенов, Данил Игоревич

AU - Ахмедов, Давид Хусенович

PY - 2025/4/29

Y1 - 2025/4/29

N2 - Path explosion is a crucial problem in symbolic execution that leads to poor performance in symbolic execution engines and hinders the widespread adoption of respective tools. A path selector is a component of a symbolic machine designed to address the path explosion problem. AI-powered path selectors have gained attention, but many challenges regarding the training process, feature selection, and information representation remain. We propose PySymGym, a framework for training AI-powered path selectors through typical supervised learning, which includes language-independent, graph-based data representation, a training protocol that minimizes manual dataset preparation, and supportive infrastructure. Evaluation of the proposed solution shows that it enables training models comparable to searchers based on manually developed heuristics: providing close coverage percentage at comparable analysis time (with the same timeout), and allowing the system to generate fewer tests.

AB - Path explosion is a crucial problem in symbolic execution that leads to poor performance in symbolic execution engines and hinders the widespread adoption of respective tools. A path selector is a component of a symbolic machine designed to address the path explosion problem. AI-powered path selectors have gained attention, but many challenges regarding the training process, feature selection, and information representation remain. We propose PySymGym, a framework for training AI-powered path selectors through typical supervised learning, which includes language-independent, graph-based data representation, a training protocol that minimizes manual dataset preparation, and supportive infrastructure. Evaluation of the proposed solution shows that it enables training models comparable to searchers based on manually developed heuristics: providing close coverage percentage at comparable analysis time (with the same timeout), and allowing the system to generate fewer tests.

KW - GNN

KW - graph neural network

KW - path selection

KW - symbolic execution

KW - training infrastructure

UR - https://www.mendeley.com/catalogue/2d255961-13a3-3385-af13-1b64fd31b4cf/

U2 - 10.1109/static66697.2025.00007

DO - 10.1109/static66697.2025.00007

M3 - Conference contribution

SN - 9798331514624

SP - 13

EP - 16

BT - 2025 IEEE/ACM 1st International Workshop on Advancing Static Analysis for Researchers and Industry Practitioners in Software Engineering (STATIC)

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 29 April 2025 through 29 April 2025

ER -