PhysicsarXiv
Ranked by formulaVerdict: NotableA Topological -Geometrical Descriptor (TGD) for Scalable and Interpretable Modeling of Two -Dimensional Carbon Networks
We introduce a compact Topological-Geometrical Descriptor (TGD) for planar two-dimensional carbon networks.
Caveats
- Preprint; not yet peer reviewed.
VerdictWorth a reader's time today.
Abstract
We introduce a compact Topological-Geometrical Descriptor (TGD) for planar two-dimensional carbon networks. Atomic coordinates are mapped to a periodic planar graph and represented through ring populations, topological-charge statistics, projected ring geometry, and dual-graph defect adjacency. The resulting modeling vector contains 19 physically interpretable features. We evaluate TGD on 120 published planar carbon allotropes using the DFT relative energy per atom with respect to graphene as the target. Repeated five-fold cross-validation with gradient boosting yields a mean absolute error (MAE) of 0.0826 +/- 0.0138 eV/atom and R2 = 0.803 +/- 0.074. Under the same validation protocol, TGD gives a mean MAE of 0.0828 eV/atom, compared with 0.0929 eV/atom for a 203-feature Matminer representation. We further analyze 360 vacancy-defected graphene configurations spanning 12 vacancy concentrations from 2 to 24 percent. When entire vacancy concentrations are held out, TGD gives a pooled MAE of 0.269 eV/vacancy and R2 = 0.861, outperforming simple baselines based on vacancy concentration alone or non-hexagonal-ring fraction alone. Sensitivity tests show that the inferred edge sets are stable to moderate changes in the adaptive bond factor. Finally, the mean absolute topological charge is examined as a one-variable empirical surrogate for the allotrope dataset and is treated explicitly as dataset-specific rather than universal. These results show that TGD combines compact prediction with direct structural interpretation while making clear the present restriction to single-layer planar carbon networks.
The editor's rubric
| Dimension | Level | Weight | What that level means |
|---|---|---|---|
| Leverage | ███░░ 3 | 18% | A method or resource many groups across the field will adopt within a year. |
| Magnitude | ███░░ 3 | 20% | Large gain: roughly 2x, or a clear new state of the art on a hard, unsaturated problem. |
| Evidence | ███░░ 3 | 22% | Solid: multiple benchmarks or cohorts, ablations, fair baselines, released code or data. |
| Novelty | ███░░ 3 | 22% | A genuinely new approach to an open problem. |
| Trajectory | ███░░ 3 | 10% | A clear path to scale. |
| Stakes | ██░░░ 2 | 8% | Benefits a professional community (practitioners, clinicians, engineers). |
Editor’s rationale
Heuristic triage from title and abstract text only, not a reading of the paper. Cues found: method (we propose); breadth (general-purpose); gains (outperforms); novelty (alternative to status quo); verification (error bars, held-out test); scale (scalable); stakes (energy).
How the score was computed
- Merit
- 5.8 / 10
- Adjusted merit
- 4.8 / 10
- Attention
- 0%
- Freshness
- 93%