{"data":{"id":"d7afe253-0d66-40c7-ba14-9a68d4f7a51a","title":"DPOison-VL: Clean Preference Poisoning in Vision-Language Direct Preference Optimization","summary":"DPOison-VL is a clean preference poisoning attack against vision-language models trained with Direct Preference Optimization (VLM-DPO). It applies imperceptible perturbations to a small share of training images, leaving prompts, responses and annotations intact, and reaches up to 98% ASR with 2% poisoned data. The authors report cross-model transferability while clean accuracy and general utility are largely preserved.","solution":"N/A -- no mitigation discussed in source.","labels":["security","research"],"sourceUrl":"https://arxiv.org/abs/2610.18609v1","publishedAt":"2026-10-13T15:58:58.000Z","cveId":null,"cweIds":null,"cvssScore":null,"cvssSeverity":null,"severity":"info","attackType":["model_poisoning"],"issueType":"research","affectedPackages":null,"affectedPackageNames":null,"affectedPackageRefs":null,"affectedVendors":[],"affectedVendorsRaw":["vision-language models","VLM-DPO"],"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":"2026-10-13T15:58:58.000Z","capecIds":null,"crossRefCount":0,"attackSophistication":"advanced","impactType":["integrity","safety"],"aiComponentTargeted":"training_data","llmSpecific":false,"classifierConfidence":0.95,"researchCategory":"preprint","atlasIds":null}}