Skip to content

task_charts__pictogram__group_difference_value

Contract

  1. Domain: charts
  2. Scene id: pictogram
  3. Public task id: task_charts__pictogram__group_difference_value
  4. Supported query_id: single

Implementation

  1. Registered class: trace_tasks.tasks.charts.pictogram.group_difference_value.ChartsPictogramGroupDifferenceValueTask
  2. Prompt bundle: src/trace_tasks/resources/prompts/charts/pictogram/charts_pictogram_v1.json
  3. Answers and annotation are produced from the same metadata execution trace.

Annotation Contract

  1. Answer schema: integer_value.
  2. Annotation schema: bbox_map.
  3. Annotation maps each compared category label to the bbox around that category row.

Program Contract

Program: difference(value(category_total(category_a, unit_scale)), value(category_total(category_b, unit_scale)), mode=absolute); output=integer_value; annotation=bbox_map(category_a_row,category_b_row); scene=pictogram; scope=group_difference_value

Candidate set: the visible pictogram rows, repeated icons, and category labels inside the group_difference_value objective scope. Operands: prompt-bound labels, categories, series names, thresholds, intervals, references, and encoded chart values, plus the task's prompt-bound target operands when present. Operation: evaluate difference over the candidate set using the filters, comparisons, aggregations, rankings, projections, or counterfactual edits named in the program expression; generation enforces a unique final answer. Output binding: answer is the integer_value value bound by integer_value. Annotation witnesses: bbox_map witnesses bound by bbox_map(category_a_row,category_b_row). Annotation maps each compared category label to the bbox around that category row. Query ids: single.

Reasoning Operations

Families: filtering, formula_evaluation