task_symbolic__braille_cell__word_braille_match_label

1) Identity

1. Domain: symbolic
2. Scene id: braille_cell
3. Scene: braille_cell
4. Task id: task_symbolic__braille_cell__word_braille_match_label
5. Objective: choose the Braille word-plate option that encodes a displayed source word.

Program Contract

Program: braille_cell.word_to_braille_match(scene=braille_cell, scope=source_word_and_four_braille_word_plate_options, output=option_label)

Candidate set: the four visible Braille word-plate option cards labeled A..D. Operands: the displayed source word and the raised-dot cell patterns shown in each option plate. Operation: encode the source word as Grade 1 Braille and select the unique option plate whose cells match that encoding. Output binding: answer is the selected option letter. Annotation witnesses: a bbox_map with source_word and selected_option roles. Query ids: single.

Reasoning Operations

Families: matching

2) Scene + Task Contract

1. Public branch metadata: query_id
2. Supported query_id: single
3. Internal prompt/query key: word_braille_match_label
4. Supported non-semantic visual axes:
5. scene_variant: clean_card|notebook_card|exam_scan
6. shared symbolic background/noise style
7. answer_gt.type: string
8. Answer schema: string, one capital-letter option label from A..D.
9. annotation_gt.type: bbox_map
10. Annotation schema: bbox_map
11. Scene contract:
12. one source word is shown,
13. four visual Braille word-plate options are labeled A..D,
14. every option plate has the same number of Braille cells as the source word length,
15. all candidate words share at least the first character, so all plate options share at least the first Braille cell,
16. exactly one option encodes the source word,
17. words use Grade 1 uncontracted lowercase Braille letters only,
18. generated words are sampled from trace_tasks.tasks.shared.word_assets,
19. word length support is 3..5.

3) Prompt Contract

1. Bundle: symbolic_braille_cell_v1
2. scene_key: braille_cell
3. task_key: word_braille_match_label
4. Required slots:
5. scene: object_description
6. answer output: json_output_contract_answer_only, answer_hint, json_example_answer_only
7. answer+annotation mode: json_output_contract, annotation_hint, answer_hint, json_example
8. Modes: answer_only, answer_and_annotation
9. Options are visual Braille word plates in the image, not prompt-only choices.

4) Annotation + Trace Contract

1. Prompt-facing annotation is a bbox_map with two semantic roles:
2. source_word: bbox around the displayed source word card,
3. selected_option: bbox around the correct Braille word-plate option.
4. Distractor option bboxes, individual Braille cells, individual dots, labels, and panel chrome are not prompt-facing annotation.
5. projected_annotation includes bbox_map and pixel_bbox_map.
6. render_map includes:
7. item_bboxes_px
8. dot_centers_px
9. raised_dot_centers_px
10. cell_dot_centers_px
11. execution_trace.braille_metadata records the source word, word length, source cell patterns, correct option label/id, option words, option cell patterns, and shared-prefix sampling metadata.

5) Determinism + Tests

1. Deterministic sampling/rendering from instance_seed.
2. Answers and annotation come from the same finalized source/option layout.
3. Behavior/trace/prompt tests: tests/test_symbolic_braille_tasks.py
4. Source-layout contract tests: tests/test_public_source_layout_contracts.py