Skip to content

task_graph__automaton__nfa_accepted_string_label

Summary

  1. Domain: graph
  2. Scene id: automaton
  3. Source package: automaton
  4. Task id: task_graph__automaton__nfa_accepted_string_label
  5. Objective: choose the candidate input string accepted by a nondeterministic finite automaton.

Query IDs

  1. nfa_accepted_string_label
  2. Query ids are internal replay metadata; public sampling is at the task-id level.

Program Contract

Program: select(candidate_label for candidate_string in options if accepts(nfa_transition_graph, candidate_string)); output=string; annotation=point_sequence(accepted_state_path); scene=automaton; scope=nfa_accepted_string_label

Candidate set: the visible graph, tree, network, route, matrix, table, node, edge, label, weight, path, and option elements inside the nfa_accepted_string_label objective scope. Operands: visible scene state and prompt-bound operands named by candidate_label, candidate_string, if, accepts, nfa_transition_graph, accepted_state_path, automaton, nfa_accepted_string_label. Operation: evaluate select over the candidate set using the visible graph structure, labels, weights, directions, reachability, paths, connectivity, or option-selection constraints encoded in the program expression; generation enforces a unique final answer. Output binding: answer uses the string schema; generation binds a unique final answer. Annotation witnesses: annotation uses the point_sequence schema; the prompt/annotation contract defines the minimal visual witnesses. Query ids: single.

Reasoning Operations

Families: topology, state_update, matching

Answer And Annotation

  1. Answer type: string.
  2. Annotation type: point_sequence.
  3. Annotation marks minimal pixel-space visual witnesses for the answer, not answer labels or non-witness annotations.
  4. Count tasks require answer_gt.value == len(annotation_gt.value) unless the annotation schema is keyed or sequence based.

Rendering Contract

  1. The scene uses the graph-domain renderer for automaton.
  2. Visual style, fonts, panel treatment, layout jitter, and post-render noise are non-semantic and must be recorded in trace metadata.
  3. Annotation projection is computed after final layout and style placement.

Prompt Contract

  1. Prompt text comes from src/trace_tasks/resources/prompts/graph/automaton/automaton_v1.json and src/trace_tasks/resources/configs/domains/graph/automaton.yaml, not hardcoded user-facing text.
  2. Answer mode emits {"answer": ...}.
  3. Answer-and-annotation mode emits {"annotation": ..., "answer": ...} with annotation matching the schema above.

Files

  1. Implementation: src/trace_tasks/tasks/graph/automaton/nfa_accepted_string_label.py
  2. Shared scene logic: src/trace_tasks/tasks/graph/automaton/shared/sampling.py
  3. Config: src/trace_tasks/resources/configs/domains/graph/automaton.yaml
  4. Prompts: src/trace_tasks/resources/prompts/graph/automaton/automaton_v1.json