{"data":{"id":"53f654c8-9f11-4278-aa10-7041391703ba","title":"From Relation-Specific Heads to Shared Interventions: Contextual Entrainment in Language Models","summary":"Researchers studied contextual entrainment, where language models reuse irrelevant tokens from earlier prompt context in later answers. Using factual-distractor relation-completion prompts across 19 lexical-semantic subtypes, they identified attention heads whose ablation reduces entrainment and tested how those head sets transfer across exact relations, semantic families, and mixed distributions. They found that same-family transfer exceeds different-family transfer and that subtype-specific searches repeatedly rediscover many globally shared heads, which can reduce harmful copying while preserving ordinary relation completion.","solution":"The paper proposes ablating a shared set of attention heads, identified using heterogeneous attack prompts, to reduce harmful context following, rather than searching for isolated relation-specific circuits.","labels":["research","security"],"sourceUrl":"https://arxiv.org/abs/2610.18674v1","publishedAt":"2026-10-13T16:33:02.000Z","cveId":null,"cweIds":null,"cvssScore":null,"cvssSeverity":null,"severity":"info","attackType":["other"],"issueType":"research","affectedPackages":null,"affectedPackageNames":null,"affectedPackageRefs":null,"affectedVendors":[],"affectedVendorsRaw":[],"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":"2026-10-13T16:33:02.000Z","capecIds":null,"crossRefCount":0,"attackSophistication":"advanced","impactType":["integrity","safety"],"aiComponentTargeted":"model","llmSpecific":true,"classifierConfidence":0.9,"researchCategory":"preprint","atlasIds":null}}