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.

Reusing Search Tree for Incremental SAT Solving of Temporal Induction

Clause Replication and Reuse in Incremental Temporal Induction

Integrating Advanced Reasoning into a SAT Solver.

On Incremental Core-Guided MaxSAT Solving

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

Explaining SAT Solving Using Causal Reasoning

A tree search algorithm towards solving Ising formulated combinatorial optimization problems

Life span of SAT techniques

Symbiosis of Search and Heuristics for Random 3-SAT

Revisiting Restarts of CDCL: Should the Search Information be Preserved?

EDA-Driven Preprocessing for SAT Solving

Between SAT and UNSAT: The Fundamental Difference in CDCL SAT

Automated Induction for Complex Data Structures

LiteSearch: Efficacious Tree Search for LLM

UNSAT Solver Synthesis via Monte Carlo Forest Search

DT-Solver: Automated Theorem Proving with Dynamic-Tree Sampling Guided by Proof-level Value Function

A Reinforcement Learning based Reset Policy for CDCL SAT Solvers

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

Chronosymbolic Learning: Efficient CHC Solving with Symbolic Reasoning and Inductive Learning

Technical Report: Enhancing LLM Reasoning with Reward-guided Tree Search

Revisiting the Learned Clauses Database Reduction Strategies