PAE Scenario Dataset: 500 Scenarios Across Customer Service, Medical Triage, Legal Advisory, and Financial Advisory

PAE Scenario Dataset 500 scenarios across four domains for evaluating peripheral context use in LLM responses. Dataset Description Each scenario contains: id: unique identifier (e.g., "cus_001", "med_001") domain: domain label (customer_service, medical, legal, finance) query: literal user query direct_answer_material: policy/guideline/reference text asker_context: entity history, profile, and patterns situational_context: temporal and environmental signals anomaly_flags: pre-computed deviations from expected patterns expert_response: gold-standard response that uses all peripheral context blind_response: technically correct response that ignores peripheral context Domains Domain Scenarios Archetypes Entity Type Customer Service 200 standard, hidden signal, red herring, multi-issue, polarity flip Customer Medical Triage 100 same, with medical-specific overrides Patient Legal Advisory 100 same, with legal-specific overrides Client Financial Advisory 100 same, with finance-specific overrides Client Scenario Archetypes (40 CS / 20 each for others) Standard: Peripheral context clearly matters; lazy agent gives worse answer Hidden signal: Subtle but important signal buried in routine details Red herring: Peripheral context exists but should not change the answer Multi-issue: 2-3 overlapping problems; context drives prioritization Polarity flip: Peripheral context reverses the surface-level answer Usage Generated by DeepSeek V4 Flash, validated by MiniMax M2.7. Used in the PAE experiment: 6 conditions, 3 model families, 4 domains Preregistered at OSF: 10.17605/OSF.IO/W3XYV Citation Lee, A. (2026). Peripheral Attention Engineering: Structured Peripheral Context Improves LLM Decision Quality. OSF Preregistration. https://doi.org/10.17605/OSF.IO/W3XYV License CC0 1.0 Universal — Public Domain Dedication