Rethinking Logic Optimization Operators: Theory-Derived Operator Compression via Agentic Source Analysis

2026-07-26

AI Summary

This work addresses the challenges in logic synthesis posed by the vast and opaque set of optimization operators, which leads to a complex search space and inefficient scheduling. For the first time, it introduces a large language model (LLM) agent that automatically infers theoretical relationships among operators from the source code semantics of ABC and mockturtle. These relationships are rigorously validated through adversarial auditing to define their applicability boundaries, enabling a deterministic gating mechanism that prunes redundant operators. The approach achieves operator compression grounded in formal relationships and exact Pareto coverage, surpassing conventional heuristic-based scheduling. Experiments demonstrate that the operator set is reduced from 40 to 31; TACO yields an 11% runtime reduction across 66 circuits with identical bit-level results; under fixed comparisons, node count and logic depth decrease by 1.0% and 3.2% on average, respectively, with a 2.6× speedup; and TACO-max attains a geometric mean NDP of 0.903 on the HeLO benchmark.

Abstract

Logic synthesis has evolved from compact two-level minimization to large multilevel flows with many interacting optimization operators. Recent work has invested substantial effort in sequencing these operators: actions are commonly treated as opaque choices in a rapidly expanding search space, while learned circuit representations and heuristic or local-greedy orchestration provide increasingly informed ways to explore it. A central obstacle is the operator vocabulary itself. Production operators are numerous, span different representations and mathematical foundations, and expose behaviors determined by implementation-level guards, bounds, and update order. We address this gap through agentic source analysis, using LLM agents to formulate operator-level relations from pinned ABC and mockturtle implementations and adversarial audits to test their stated scope. The resulting certified relations yield theory-derived operator compression: 40 deployed recipe actions collapse to a 31-action exact Pareto cover, and source-level conditions compile into deterministic admission gates. We integrate these gates directly into ABC Orchestrate to form TACO. Two exact gates reduce Orchestrate runtime by 11% with bit-identical outputs on 66 circuits. In a held-fixed integrated comparison, TACO uses fewer nodes on 14 of 16 circuits, with geometric-mean reductions of 1.0% in nodes and 3.2% in levels, while running 2.6x faster. TACO-max achieves an NDP geometric-mean ratio of 0.903 on HeLO's three exact-input rows.

Problem

• logic synthesis
• optimization operators
• operator compression
• agentic analysis
• circuit optimization

Innovation

• agentic source analysis
• operator compression
• logic synthesis
• Pareto cover
• deterministic admission gates