LLM4DV Framework

Tool

First seen 7/5/2026

Last seen 7/5/2026

Evidence 6 chunks

NEIGHBORHOOD

No graph connections found for this entity yet. It may appear in future ingestion runs.

18 connections

Coverage Plan uses → 95% 2e

The LLM4DV framework uses coverage plans to guide the stimulus generation process.

The LLM4DV framework uses dialogue restarting as a prompting improvement.

The LLM4DV framework uses GPT-3.5-turbo-0613 in fixed-budget experiments.

The LLM4DV framework uses missed-bin sampling as a prompting improvement.

The LLM4DV framework uses best-iterative-message sampling as a prompting improvement.

Verilator uses → 97% 1e

The LLM4DV framework uses Verilator for simulation and testing of DUT modules.

CocoTB uses → 97% 1e

The LLM4DV framework uses cocotb for simulation and testing of DUT modules.

Llama 2 7B uses → 95% 1e

The LLM4DV framework also uses Llama 2 7B model in ablation experiments.

The LLM4DV framework is evaluated on the Primitive Data Prefetcher Core DUT module.

The LLM4DV framework is evaluated on the Ibex Instruction Decoder DUT module.

Ibex CPU evaluates → 98% 1e

The LLM4DV framework is evaluated on the Ibex CPU DUT module.

Test Stimuli Generation implements → 97% 1e

The LLM4DV framework implements LLM-based test stimuli generation for hardware designs.

Design Under Test uses → 95% 1e

The LLM4DV framework uses DUT modules as its test targets.

Coverage Bin uses → 95% 1e

The LLM4DV framework tracks coverage bins to monitor progress.

LLM4DV ← introduces 99% 1e

The paper LLM4DV presents and introduces the LLM4DV benchmarking framework.

RTL uses → 85% 1e

The LLM4DV framework applies LLMs to hardware designs described in HDL/RTL.

The LLM4DV framework utilizes LLMs to generate test stimuli for hardware designs.

The LLM4DV framework uses the Coverage-Feedback Template to generate prompts for the LLM.