Component ablation — is the architecture load-bearing?
Beating weak baselines is not enough; we must show the compiler's architecture carries the result. Each row removes exactly one component and keeps the other five intact, then scores on the identical rubric. A large drop in composite from a removal means that component is load-bearing. If removing a component changes nothing, it was decoration.
orkg_biomedicine_biomedical_research_in_psychiatr
| Configuration | Composite | Δ vs full | Grounding | Falsifiable | Rediscovery |
|---|---|---|---|---|---|
| Full compiler | 0.98 | baseline | 1.00 | 1.00 | 0.00 |
| No graph reasoning | 0.00 | -0.976 | 0.00 | 1.00 | 0.00 |
| No grounding verification | 0.98 | +0.000 | 1.00 | 1.00 | 0.00 |
| No gap detection | 0.78 | -0.195 | 1.00 | 1.00 | 0.00 |
| No falsifiability eval | 0.51 | -0.463 | 1.00 | 0.00 | 0.00 |
| No literature synthesis | 0.91 | -0.066 | 0.88 | 1.00 | 0.00 |
orkg_chemistry_chemical_sensors
| Configuration | Composite | Δ vs full | Grounding | Falsifiable | Rediscovery |
|---|---|---|---|---|---|
| Full compiler | 0.99 | baseline | 1.00 | 1.00 | 0.00 |
| No graph reasoning | 0.00 | -0.987 | 0.00 | 1.00 | 0.00 |
| No grounding verification | 0.99 | -0.000 | 1.00 | 1.00 | 0.00 |
| No gap detection | 0.98 | -0.004 | 1.00 | 1.00 | 0.00 |
| No falsifiability eval | 0.52 | -0.472 | 1.00 | 0.00 | 0.00 |
| No literature synthesis | 0.82 | -0.165 | 0.58 | 1.00 | 0.00 |
orkg_computer_science_semantic_segmentation
| Configuration | Composite | Δ vs full | Grounding | Falsifiable | Rediscovery |
|---|---|---|---|---|---|
| Full compiler | 0.68 | baseline | 0.35 | 1.00 | 0.00 |
| No graph reasoning | 0.00 | -0.684 | 0.00 | 1.00 | 0.00 |
| No grounding verification | 0.99 | +0.305 | 1.00 | 1.00 | 0.00 |
| No gap detection | 0.91 | +0.225 | 0.80 | 1.00 | 0.00 |
| No falsifiability eval | 0.52 | -0.165 | 1.00 | 0.00 | 0.00 |
| No literature synthesis | 0.89 | +0.201 | 0.75 | 1.00 | 0.00 |
orkg_environmental_science_biodiversity_inventories_with_dn
| Configuration | Composite | Δ vs full | Grounding | Falsifiable | Rediscovery |
|---|---|---|---|---|---|
| Full compiler | 0.98 | baseline | 1.00 | 1.00 | 0.00 |
| No graph reasoning | 0.00 | -0.977 | 0.00 | 1.00 | 1.00 |
| No grounding verification | 0.98 | +0.005 | 1.00 | 1.00 | 0.20 |
| No gap detection | 0.88 | -0.096 | 0.79 | 1.00 | 1.00 |
| No falsifiability eval | 0.52 | -0.460 | 1.00 | 0.00 | 1.00 |
| No literature synthesis | 0.98 | +0.000 | 1.00 | 1.00 | 1.00 |
orkg_materials_science_mapping_precursor_chemistries_us
| Configuration | Composite | Δ vs full | Grounding | Falsifiable | Rediscovery |
|---|---|---|---|---|---|
| Full compiler | 0.89 | baseline | 0.75 | 1.00 | 0.00 |
| No graph reasoning | 0.00 | -0.892 | 0.00 | 1.00 | 0.00 |
| No grounding verification | 0.98 | +0.090 | 1.00 | 1.00 | 0.00 |
| No gap detection | 0.78 | -0.113 | 1.00 | 1.00 | 0.00 |
| No falsifiability eval | 0.52 | -0.376 | 1.00 | 0.00 | 0.00 |
| No literature synthesis | 0.97 | +0.081 | 1.00 | 1.00 | 0.00 |
orkg_neuroscience_simultaneous_localization_and_ma
| Configuration | Composite | Δ vs full | Grounding | Falsifiable | Rediscovery |
|---|---|---|---|---|---|
| Full compiler | 0.85 | baseline | 0.70 | 1.00 | 0.00 |
| No graph reasoning | 0.00 | -0.851 | 0.00 | 1.00 | 0.00 |
| No grounding verification | 0.98 | +0.132 | 1.00 | 1.00 | 0.00 |
| No gap detection | 0.98 | +0.130 | 1.00 | 1.00 | 0.00 |
| No falsifiability eval | 0.50 | -0.352 | 0.92 | 0.00 | 0.00 |
| No literature synthesis | 0.99 | +0.138 | 1.00 | 1.00 | 0.00 |
orkg_physics_dataset_used_in_wind_energy_pote
| Configuration | Composite | Δ vs full | Grounding | Falsifiable | Rediscovery |
|---|---|---|---|---|---|
| Full compiler | 0.88 | baseline | 0.75 | 1.00 | 0.00 |
| No graph reasoning | 0.00 | -0.884 | 0.00 | 1.00 | 0.00 |
| No grounding verification | 0.98 | +0.098 | 1.00 | 1.00 | 0.00 |
| No gap detection | 0.98 | +0.099 | 1.00 | 1.00 | 0.00 |
| No falsifiability eval | 0.52 | -0.366 | 1.00 | 0.00 | 0.00 |
| No literature synthesis | 0.98 | +0.097 | 1.00 | 1.00 | 0.00 |
Ablations run on one representative domain per discipline. Red Δ = large degradation = that component is genuinely load-bearing.