{"data":{"id":"dfa88e14-7395-4ed8-bb6b-52adac46ca1e","title":"AfterDeleteBench: Auditing Post-Deletion Information Persistence in LLM Agents","summary":"AfterDeleteBench is a controlled behavioral audit that tests whether LLM agents still reveal information after its source memory record is deleted. In the primary setting, one query recovered a deleted synthetic secret from Qwen2.5-7B-Instruct in roughly two thirds of cases, while matched never-exposed controls showed zero recovery. Reconstruction succeeded in 62-99% of cases across Qwen, Mistral, and Phi-3 when only separately stored components remained, and provenance-aware closure eliminated observed recovery in the tested settings.","solution":"Provenance-aware closure eliminated observed recovery in the tested positive-baseline settings. An exact-literal write filter blocks reinsertion but cannot erase information already present.","labels":["security","research"],"sourceUrl":"https://arxiv.org/abs/2610.31366v1","publishedAt":"2026-10-22T14:41:36.000Z","cveId":null,"cweIds":null,"cvssScore":null,"cvssSeverity":null,"severity":"info","attackType":["data_extraction","pii_leakage"],"issueType":"research","affectedPackages":null,"affectedPackageNames":null,"affectedPackageRefs":null,"affectedVendors":[],"affectedVendorsRaw":["Qwen2.5-7B-Instruct","Mistral","Phi-3"],"classifierModel":"claude-haiku-5-5","classifierPromptVersion":"v4","summaryPromptVersion":"v2","headline":null,"headlinePromptVersion":null,"cvssVector":null,"attackVector":null,"attackComplexity":null,"privilegesRequired":null,"userInteraction":null,"exploitMaturity":null,"epssScore":null,"epssCheckedAt":null,"kevDateAdded":null,"advisoryAliases":null,"affectedPackagesSource":null,"affectedPackagesCheckedAt":null,"patchAvailable":null,"disclosureDate":"2026-10-22T14:41:36.000Z","capecIds":null,"crossRefCount":0,"attackSophistication":"moderate","impactType":["confidentiality"],"aiComponentTargeted":"agent","llmSpecific":true,"classifierConfidence":0.95,"researchCategory":"preprint","atlasIds":null}}