Abstract:The past three decades have witnessed notable success in designing efficient SAT solvers, with modern solvers capable of solving industrial benchmarks containing millions of variables in just a few seconds. The success of modern SAT solvers owes to the widely-used CDCL algorithm, which lacks comprehensive theoretical investigation. Furthermore, it has been observed that CDCL solvers still struggle to deal with specific classes of benchmarks comprising only hundreds of variables, which contrasts with their widespread use in real-world applications. Consequently, there is an urgent need to uncover the inner workings of these seemingly weak yet powerful black boxes. In this paper, we present a first step towards this goal by introducing an approach called CausalSAT, which employs causal reasoning to gain insights into the functioning of modern SAT solvers. CausalSAT initially generates observational data from the execution of SAT solvers and learns a structured graph representing the causal relationships between the components of a SAT solver. Subsequently, given a query such as whether a clause with low literals blocks distance (LBD) has a higher clause utility, CausalSAT calculates the causal effect of LBD on clause utility and provides an answer to the question. We use CausalSAT to quantitatively verify hypotheses previously regarded as "rules of thumb" or empirical findings such as the query above. Moreover, CausalSAT can address previously unexplored questions, like which branching heuristic leads to greater clause utility in order to study the relationship between branching and clause management. Experimental evaluations using practical benchmarks demonstrate that CausalSAT effectively fits the data, verifies four "rules of thumb", and provides answers to three questions closely related to implementing modern solvers.

Integrating Advanced Reasoning into a SAT Solver.

Explaining SAT Solving Using Causal Reasoning

EHSAT: an Efficient RTL Satisfiability Solver Using an Extended DPLL Procedure.

SAT Solver with Static and Dynamic Ordering Decision Making

An Efficient Resolution Based Algorithm for SAT

A Folding Strategy for SAT Solvers Based on Shannon's Expansion Theorem

Solving SAT Problem by Heuristic Polarity Decision-Making Algorithm

EDA-Driven Preprocessing for SAT Solving

AutoSAT: Automatically Optimize SAT Solvers via Large Language Models

SatLM: Satisfiability-Aided Language Models Using Declarative Prompting

When Do Program-of-Thought Works for Reasoning?

Improving abcdSAT by At-Least-One Recently Used Clause Management Strategy

Accelerating an FPGA-Based SAT Solver by Software and Hardware Co-design

From Total Assignment Enumeration to Modern SAT Solver

New reasoning algorithm based on EFALC

Lightweight Online Learning for Sets of Related Problems in Automated Reasoning

Enhance SAT Conflict Analysis for Model Checking

Reasoning Within Extended Fuzzy Description Logic Supporting Terminological Axiom Restrictions

RTSAT: A Hybrid Satisfiability Solver for RTL Circuits

Reusing Search Tree for Incremental SAT Solving of Temporal Induction