discover is good because it asks coverage questions. Soft intent framing,
multi-select scope sweeps, named alternatives in non-goals. Half of those
questions have no downstream parser; they exist for the operator’s own
framing. They’re what keep the FRD on the ask you actually made.
We A/B’d it against an adaptation of SAGE-Agent (arXiv 2511.08798), which replaces the question tree with EVPI scoring over a small candidate set and a principled stop signal. 16 cases stratified across task type × codebase familiarity, two blind LLM judges (Opus + Sonnet), pre-registered 7-dimension rubric. Inter-judge Pearson 0.696, exact-match 81%, within-one 100%.
Scores (joint, mean across both judges, out of 21). The W/L/D column is the variant’s head-to-head record against discover across all 64 pair-comparisons per dimension (8 FRDs × 8 FRDs):
| Dimension | discover | SAGE variant | variant vs discover W/L/D | Effect |
|---|---|---|---|---|
| D1 Completeness | 3.00 | 2.75 | 0 / 24 / 40 | medium |
| D2 Specificity | 2.88 | 2.69 | 6 / 21 / 37 | small |
| D3 No-impl-leak | 2.38 | 2.56 | 31 / 16 / 17 | small (variant) |
| D4 Anti-rescoping | 2.81 | 2.00 | 2 / 51 / 11 | large |
| D5 Consistency | 2.94 | 2.94 | 7 / 7 / 50 | tie |
| D6 Auditability | 2.56 | 2.19 | 15 / 33 / 16 | small |
| D7 Actionability | 2.69 | 2.44 | 16 / 29 / 19 | small |
| Total | 19.26 | 17.57 | 77 / 181 / 190 of 448 | discover +1.69 |
Six dimensions favor discover, one ties, one (D3) favors the variant.
The deciding dimension is D4, anti-rescoping discipline, by a large
effect. EVPI’s turn-saving mechanism skips exactly the comprehensive
scope-coverage sweeps that drive D4. Optimizing for “smallest question
set to disambiguate intent” turns out to be the wrong objective when the
artifact is a requirements document; the right objective is “did we
consider what I’m leaving out?”
We’re keeping discover. The variant’s candidate-shape framing and
Belief Trace audit block are worth porting back as a smaller change,
without the EVPI scoring that suppressed the sweeps in the first place.