Backus-Naur Form Grammar Research Articles

Enhancing Program Synthesis with Large Language Models Using Many-Objective Grammar-Guided Genetic Programming

The ability to automatically generate code, i.e., program synthesis, is one of the most important applications of artificial intelligence (AI). Currently, two AI techniques are leading the way: large language models (LLMs) and genetic programming (GP) methods—each with its strengths and weaknesses. While LLMs have shown success in program synthesis from a task description, they often struggle to generate the correct code due to ambiguity in task specifications, complex programming syntax, and lack of reliability in the generated code. Furthermore, their generative nature limits their ability to fix erroneous code with iterative LLM prompting. Grammar-guided genetic programming (G3P, i.e., one of the top GP methods) has been shown capable of evolving programs that fit a defined Backus–Naur-form (BNF) grammar based on a set of input/output tests that help guide the search process while ensuring that the generated code does not include calls to untrustworthy libraries or poorly structured snippets. However, G3P still faces issues generating code for complex tasks. A recent study attempting to combine both approaches (G3P and LLMs) by seeding an LLM-generated program into the initial population of the G3P has shown promising results. However, the approach rapidly loses the seeded information over the evolutionary process, which hinders its performance. In this work, we propose combining an LLM (specifically ChatGPT) with a many-objective G3P (MaOG3P) framework in two parts: (i) provide the LLM-generated code as a seed to the evolutionary process following a grammar-mapping phase that creates an avenue for program evolution and error correction; and (ii) leverage many-objective similarity measures towards the LLM-generated code to guide the search process throughout the evolution. The idea behind using the similarity measures is that the LLM-generated code is likely to be close to the correct fitting code. Our approach compels any generated program to adhere to the BNF grammar, ultimately mitigating security risks and improving code quality. Experiments on a well-known and widely used program synthesis dataset show that our approach successfully improves the synthesis of grammar-fitting code for several tasks.

A high-level domain-specific language for SIEM (design, development and formal verification)

Organizations deploy security information and event management (SIEM) systems for centralized management of security events. The real-time security monitoring capability of the SIEM depends on the correlation process where events data are matched against the security rules. Most SIEM systems use general purpose languages to define security rules. Creating new rules in general purpose languages require excellent programming skills in the proprietary language and intimate knowledge of events. This paper introduces a high-level domain-specific language (HDSL) which simplifies rule creation for the SIEM system. We formally specify the HDSL with extended Backus–Naur form grammar in another tool for language recognition according to the model driven engineering approach. In our implementation framework, the rules defined in the HDSL are converted in the standard event processing language. For evaluation purpose, the converted security rules are tested on the service real-time data security analytics. The results indicate that the rules are converted accurately and generate alarms when specific attacks are detected. For checking correctness of the HDSL, formal verification is carried out using satisfiability modulo theory and Z3 solver. The results are evaluated under diverse attack scenarios, which reveal that HDSL is functioning correctly. The HDSL enhances the SIEM correlation capabilities by providing a tranquil approach for writing the correlation rules.

Grammatical evolution using two-dimensional gene for symbolic regression: an advanced improvement with conditional statement grammar

Symbolic regression problems can be solved using grammatical evolution (GE), an evolutionary computation (EC) method, to find a function that coincides satisfactorily with the given datasets. The evolutional approach of GE is based on the grammar learning paradigm, which can translate the genotype (binary digit) into the phenotype (terminals and non-terminals). Unlike traditional codons in a genotype, the fittest codons in phenotype represented by the Backus-Naur form (BNF) are difficult for next generation genes to inherit the traits of parents, accounting for crossover and mutation. For this issue, this article presents a proposal of an advanced improvement to GE using a two-dimensional gene (GE2DG). In contrast to multi-chromosomal GE (GEMC), our proposal not only encloses the two-dimensional gene-expression for symbolic regression, but also introduces one independent gene defined as a conditional statement to express a new BNF grammar of an if-then (-else) branch. In the experiments described herein, continuous/discontinuous non-branch functions and continuous/discontinuous branch functions, four testing patterns, are considered as numerical examples. Results show that GE2DG has better performance than the original GE or GEMC. Especially for the case of branch functions, GE with hybrid chromosome (GEHC), where GE2DG is incorporated with GEMC, has faster convergence in symbolic regression than other methods.

Grammatical Swarm: The generation of programs by social programming

This study examines Social Programming, that is, the construction of programs using a Social Swarm algorithm based on Particle Swarm Optimization. Each individual particle represents choices of program construction rules, where these rules are specified using a Backus–Naur Form grammar. This study represents the first instance of a Particle Swarm Algorithm being used to generate programs. A selection of benchmark problems from the field of Genetic Programming are tackled and performance is compared to Grammatical Evolution. The results demonstrate that it is possible to successfully generate programs using the Grammatical Swarm technique. An analysis of the Grammatical Swarm approach is presented on the dynamics of the search. It is found that restricting the search to the generation of complete programs, or with the use of a ratchet constraint forcing individuals to move only if a fitness improvement has been found, can have detrimental consequences for the swarms performance and dynamics.

Auditor efficiency and effectiveness with diagrammatic and linguistic conceptual model representations

The differences in efficiency, effectiveness, ease of use, and user satisfaction among auditors using diagrammatic and linguistic conceptual model representations were investigated. The diagrammatic conceptual model was the entity–relationship (ER) model [Batini C, Ceri S, Navathe SB. Conceptual database design: an entity relationship approach. Redwood City, CA: Benjamin/Cummings, 1992.]. The linguistic conceptual model was an informationally equivalent Backus–Naur Form (BNF) grammar (as described in Batini et al., 1992). The conceptual model representation was manipulated in an experiment containing three types of tasks: search, recognition, and inference. Participants in the study were 46 first year auditors employed by a Big 5 accounting firm, who were enrolled in an accounting information systems course in a graduate accounting degree program. Results indicate that participants considered the ER model easier to use, used it faster, and were more satisfied with it, than the BNF grammar. However, there were no significant differences between representations with respect to accuracy.