Typed decision models need an interface that can use additional reasoning while retaining candidate outputs.
We train a shared candidate-answer reader on a frozen Qwen3.5-4B backbone to unify direct decisions with on-demand, interruptible reasoning.
The reader returns distributions from empty and partial reasoning prefixes.
A benefit gate starts reasoning, a natural-state trigger requests explicit reads, and an entropy rule accepts acquired answers and cancels generation.
Measurements distinguish answer-query work, generated output, and pipeline time.
Accuracy Improvements
Across 768 questions from ProofWriter, LogiQA 2.0, and ARC-Challenge, accuracy rises from 70.96% for direct reading to 84.38%.
Relative to 512-token terminal reading, accuracy improves by 7.29 percentage points with 2.25% more mean output work.
Relative to 1,024-token terminal reading, pipeline time falls by 46.30% with an accuracy difference of -1.17 points.
Controlled Experiments
Exact-prefix controls isolate the saving from interruption, while read-scheduling controls distinguish reduced observation work from shorter response time.