Published July 16, 2026
| Version v1
Dataset
Open
Replication Package for 'To Think or Not to Think: Evaluating LLM Reasoning and Agents in Vulnerability Detection'
Authors/Creators
Description
PrimeVul-Pair vuln-detection experiments across 4 models × 2 modes × 2 prompting × 4 agent designs; contains dataset, run scripts (NA/SA/DA/MA + budget baselines), raw outputs, metrics/cost/energy, RQ3 explanation-quality rater outputs.