repro-formal-problem-solving / outputs /claim5 /claim5_native_focused.json
ProCreations's picture
Repair claim 5 with native benchmark execution
35042ba
Raw
History Blame Contribute Delete
2.94 kB
{
"protocol": {
"paper": "arXiv:2509.22259v1, Section 4.1 and Appendix A.1",
"official_repository": "https://github.com/cederikhoefs/Graph-RoPE",
"official_commit": "4ac067eb38272543b0cdd7591d630399ff37bce4",
"graph_sizes": {
"monochromatic": "5x5 grid",
"shortest": "10-node Watts-Strogatz, k=2, p=0.6"
},
"model": "4-layer single-head transformer, hidden=32, MLP=32, dropout=0.2",
"optimizer": "AdamW(lr=2e-4, weight_decay=1e-4), cosine eta_min=2e-6",
"omega_initialization": "zero",
"budget": {
"train_examples": 256,
"test_examples": 128,
"epochs": 8,
"seeds": [
0,
1
]
},
"normalization": "RMSE divided by graph size (25 for monochromatic, 10 for shortest path)",
"baseline": "m=0, same model and data with WIRE disabled"
},
"rows": [
{
"task": "monochromatic",
"setting": "0",
"m": 0,
"seed": 0,
"train_examples": 256,
"test_examples": 128,
"epochs": 8,
"best_normalized_test_rmse": 0.06839866191148758,
"parameters": 26369
},
{
"task": "monochromatic",
"setting": "0",
"m": 0,
"seed": 1,
"train_examples": 256,
"test_examples": 128,
"epochs": 8,
"best_normalized_test_rmse": 0.06361407786607742,
"parameters": 26369
},
{
"task": "monochromatic",
"setting": "0",
"m": 5,
"seed": 0,
"train_examples": 256,
"test_examples": 128,
"epochs": 8,
"best_normalized_test_rmse": 0.07993544638156891,
"parameters": 26689
},
{
"task": "monochromatic",
"setting": "0",
"m": 5,
"seed": 1,
"train_examples": 256,
"test_examples": 128,
"epochs": 8,
"best_normalized_test_rmse": 0.068186916410923,
"parameters": 26689
},
{
"task": "monochromatic",
"setting": "15",
"m": 0,
"seed": 0,
"train_examples": 256,
"test_examples": 128,
"epochs": 8,
"best_normalized_test_rmse": 0.062019895762205124,
"parameters": 26369
},
{
"task": "monochromatic",
"setting": "15",
"m": 0,
"seed": 1,
"train_examples": 256,
"test_examples": 128,
"epochs": 8,
"best_normalized_test_rmse": 0.06363179534673691,
"parameters": 26369
},
{
"task": "monochromatic",
"setting": "15",
"m": 5,
"seed": 0,
"train_examples": 256,
"test_examples": 128,
"epochs": 8,
"best_normalized_test_rmse": 0.055608343333005905,
"parameters": 26689
},
{
"task": "monochromatic",
"setting": "15",
"m": 5,
"seed": 1,
"train_examples": 256,
"test_examples": 128,
"epochs": 8,
"best_normalized_test_rmse": 0.06651563197374344,
"parameters": 26689
}
],
"runtime_seconds": 127.14733576774597
}