English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

EquivSVA: A Formally Verified Dataset of Behavioral SystemVerilog Assertion Families for Robustness Research

Forum topic · 小凯 · 2026-09-24

Summary

EquivSVA (arXiv:2609.26751) is a formally verified dataset for studying whether LLM-generated SystemVerilog Assertions capture externally observable behavior rather than incidental details of a single RTL implementation. It organizes 120 behavior families across 12 categories, each containing four structurally distinct but equivalent RTL implementations, shared interface-level gold properties (914 total), three controlled mutants (360 total), and formal-validation evidence. Every family passes a fixed 17-job validation suite covering RTL equivalence, gold-property proofs, property reachability, mutant distinguishability, and gold-property checks on mutants, with fixed family-safe train/dev/test splits. As a demonstration, the authors evaluated Qwen2.5-Coder-7B-Instruct on the held-out test split: of 293 interface-only generated properties, 93 were formally sound, and soundness varied across equivalent implementations for 14 of 24 test families—showing that behavior-family organization enables controlled robustness studies of assertion generation without changing intended functionality. Data and tools are released at https://github.com/aditigupta96/EquivSVA.

Paper Overview

Field: Chip/Hardware Verification Author: FNU Aditi Published: 2026-09-22 arXiv: 2609.26751

Summary

Large language models are increasingly used to generate SystemVerilog Assertions from natural-language specifications and register-transfer-level designs. Existing datasets and benchmarks support important goals such as large-scale training, formal evaluation, specification-to-assertion generation, and mutation-based testing. A complementary need is to study whether a generated assertion captures externally observable behavior or depends on incidental details of one RTL implementation.

EquivSVA is a formally verified dataset organized around behavior families. Each family contains four structurally distinct RTL implementations of the same externally observable behavior, shared interface-level gold properties, three controlled mutants, and formal-validation evidence.

Dataset Contents

  • 120 behavior families across 12 categories
  • 480 reference RTL implementations
  • 914 gold properties
  • 360 mutants
  • Fixed family-safe train, development, and test splits
  • Every final family passes a fixed 17-job validation suite covering:

    1. RTL equivalence 2. Gold-property proofs 3. Property reachability 4. Mutant distinguishability 5. Gold-property checks on mutants

    Case Study

    As a small demonstration of the analyses enabled by the dataset, the authors evaluate the publicly released, Apache-2.0-licensed Qwen2.5-Coder-7B-Instruct model on the held-out test split:

  • Of 293 interface-only generated properties, 93 are formally sound
  • The number of sound properties varies across equivalent implementations for 14 of 24 test families
These results illustrate how behavior-family organization can support controlled studies of assertion-generation robustness without requiring changes in intended functionality.

Resources

The dataset, generators, validation scripts, and case-study artifacts are publicly released at:

https://github.com/aditigupta96/EquivSVA

Tags

#formal-verification#systemverilog-assertions#llm#rtl-design#chip-verification#datasets#arxiv#mutation-testing

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178635151