JasonYan777/PersonaSignal-PerceivabilityTest-Planning-Horizon-gpt-4o-mini

Dataset card for PersonaSignal-PerceivabilityTest-Planning-Horizon-gpt-4o-mini

This dataset was made with Curator.

Dataset details

A sample from the dataset:

{
    "dimension_name": "planning_horizon",
    "dimension_values": [
        "Spontaneous",
        "Balanced",
        "Strategic"
    ],
    "dimension_description": "Captures how the user sequences work and values payoff timing. Spontaneous favors immediate action, short feedback loops, and minimal upfront planning. Balanced outlines a short sequence with a checkpoint and simple contingencies. Strategic frames a long run objective with phased milestones, leading indicators, and tolerance for delayed payoff.",
    "question": "I want to get back into running after a long break. How should I structure the first month?",
    "why_differ": "Spontaneous will emphasize quick starts with frequent, short sessions and immediate feel-based adjustments. Balanced will propose a simple weekly outline with a mid-point check. Strategic will outline a progressive build oriented toward later consistency and endurance markers.",
    "how_subtle": "Differences show up in session cadence, whether a mid-month review appears, and whether goals refer to immediate comfort vs longer capability, not in overt time-horizon words.",
    "sampled_value": "Strategic",
    "num_distractors": 5,
    "ground_truth_persona": "A runner returning after a long break who wants a phased plan tied to a longer-term goal beyond the first month. Prefers clear weekly progression, a recovery week, and measurable indicators like total time and perceived effort trends. Comfortable delaying harder sessions until base metrics stabilize and wants milestones that set up future race training.",
    "distractor_personas": [
        "Wants to start immediately without a fixed calendar, choosing run length and pace based on daily feel. Prefers quick feedback and adjusting day to day rather than following a set sequence.",
        "Motivated by momentum and short wins, they\u2019d rather keep plans loose and decide on the spot. Open to shifting or skipping days as needed instead of sticking to a predefined schedule.",
        "Wants a simple 4-week outline with a mid-month check-in and basic adjustments if needed. Prefers a modest weekly build and clear rest days, without heavy analysis.",
        "Wants each week laid out with targets, plus a brief review after two weeks to tweak volume. Comfortable with a light structure and simple contingencies if a day is missed.",
        "Prefers a basic month plan divided into weeks, each with a focus and a short end-of-week reflection. Values steady, step-by-step progress without deep metrics."
    ],
    "personalized_response": "To ease back into running effectively, a structured plan for your first month can help you build a solid foundation and avoid injury. Here's a suggested week-by-week approach that includes a mix of running, walking, and rest days, as well as a recovery week:\n\n### Week 1: Getting Started\n- **Goal**: Establish a routine and assess comfort levels.\n- **Days**: 3 running days, 2 cross-training days, 2 rest days.\n- **Running Sessions**:\n  - **Session 1**: 15 min walk + 5 min run (easy pace) + 15 min walk\n  - **Session 2**: 20 min walk + 10 min run (easy pace) + 10 min walk\n  - **Session 3**: 15 min easy jog + 10 min walk\n- **Cross-Training**: Activities like cycling or swimming for 20-30 min.\n- **Total Time**: Aim for about 45-60 min per session.\n",
    "correct_choice": "E",
    "judge_choice": "E",
    "judge_rationale": "The response provides a structured month-long running plan with specific week-by-week details, which aligns with the persona E's preference for a phased approach tied to longer-term goals. It emphasizes building a solid foundation, tracking progress, and includes a recovery week, suggesting that the user is looking for a clear progression that prepares them for future running challenges. This detailed guidance indicates an awareness of the needs of someone returning to running after a break, which is characteristic of persona E.",
    "reward": 1
}

Loading the dataset

You can load this dataset using the following code:

from datasets import load_dataset

dataset = load_dataset("JasonYan777/PersonaSignal-PerceivabilityTest-Planning-Horizon-gpt-4o-mini")