pith:VH5WU2S2
Physics Guided Generative Optimization for Trotter Suzuki Decomposition
A conditional diffusion model guided by physics-informed fidelity feedback produces Trotter-Suzuki decompositions that reach 85.6 percent of fourth-order baseline accuracy at 22 percent circuit depth on the transverse Ising model.
arxiv:2605.13268 v1 · 2026-05-13 · quant-ph · cs.LG
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{VH5WU2S2IPZA6JEYHSNZLKJRVL}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
On the transverse field Ising model under the primary comparison setup, the method reaches 85.6% of the fidelity of a fourth order Qiskit baseline (0.856) at roughly 21.8% of the circuit depth and 19.2% of the baseline CNOT count. Under an equal depth budget, fine tuning in the loop reached a best observed fidelity of 0.9994.
The physics-informed neural network supplies reliable differentiable fidelity feedback that remains accurate across the discrete grouping and order choices explored during training and that the REINFORCE-based optimization converges to strategies that generalize beyond the specific TFIM instances and hyperparameter settings used.
A generative optimization loop using diffusion models, PINNs, and GNNs achieves 85.6% of fourth-order Qiskit fidelity at 21.8% circuit depth for transverse-field Ising model Trotter-Suzuki decomposition.
References
Receipt and verification
| First computed | 2026-05-18T02:44:49.298575Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
a9fb6a6a5a43f20f24983c9b95a931aad56a6f83af2f9978bbc83f647d484a69
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/VH5WU2S2IPZA6JEYHSNZLKJRVL \
| jq -c '.canonical_record' \
| python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: a9fb6a6a5a43f20f24983c9b95a931aad56a6f83af2f9978bbc83f647d484a69
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "5df33a0ff3f8235550387f506e2792a0f0ca48fa1e73275d70403ebbdacd6cd2",
"cross_cats_sorted": [
"cs.LG"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"primary_cat": "quant-ph",
"submitted_at": "2026-05-13T09:48:33Z",
"title_canon_sha256": "c379176237dc6f6c7746dc639c7dd1f0cb92e932b0e51e0f28f9c1dd83946d67"
},
"schema_version": "1.0",
"source": {
"id": "2605.13268",
"kind": "arxiv",
"version": 1
}
}