SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2603.06260 · 2026

Learning to Solve Orienteering Problem with Time Windows and Variable Profits

Marco Roveri, Zanxi Ruan, Luigi Palopoli, Songqun Gao, Patrick Floor, Daniele Fontanelli

arXiv · PDF · Open in the Atlas

Code that ran

4 of 7 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (3 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 3 have no contract verdict; 1 failure of our instrument; 3 no recorded run), measured 2026-11-22; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2603.06260.json · badge: /api/ran/2603.06260.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
SwonGao/DeCoST community
Papers with Code archive
4 ran · 3 no recorded run
FunctionStatusWhere it lives
GAT Ran
ran on a synthesized input · constructed an object; no result was computed
SwonGao/DeCoST/DeCoST/models/SINGLEModel.py
code served (permissive licence) · get_code("25f967655d033233")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, GAT: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "in_features": 8,
    "n_hidden": 16,
    "n_heads": 2,
    "num_classes": 3,
    "concat": False,
    "dropout": 0.4,
    "leaky_relu_slope": 0.2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type GAT — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a GAT output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
GraphAttentionLayer Ran
ran on a synthesized input · constructed an object; no result was computed
SwonGao/DeCoST/DeCoST/models/SINGLEModel.py
code served (permissive licence) · get_code("1ed80a36e25591db")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, GraphAttentionLayer: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Define minimal setup for GraphAttentionLayer
batch_size = 2
n_nodes = 4
in_features = 3
out_features = 8
n_heads = 2

# Create input tensors
h = torch.randn(batch_size, n_nodes, in_features)

# Create adjacency matrix (sparse, with some connections)
adj_mat = torch.zeros(n_heads, n_nodes, n_nodes)
# Add some connections
adj_mat[:, 0, 1] = 1
adj_mat[:, 1, 0] = 1
adj_mat[:, 1, 2] = 1
adj_mat[:, 2, 1] = 1
adj_mat[:, 2, 3] = 1
adj_mat[:, 3, 2] = 1
# Add self-loops
for i in range(n_nodes):
    adj_mat[:, i, i] = 1

ARGS = {
    'in_features': in_features,
    'out_features': out_features,
    'n_heads': n_heads,
    'concat': True,
    'dropout': 0.4,
    'leaky_relu_slope': 0.2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type GraphAttentionLayer — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a GraphAttentionLayer output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
SINGLE_Decoder Ran
ran on a synthesized input · constructed an object; no result was computed
SwonGao/DeCoST/DeCoST/models/SINGLEModel.py
code served (permissive licence) · get_code("3687fd25dac37a34")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, SINGLE_Decoder: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'problem': 'CVRP',
    'embedding_dim': 4,
    'head_num': 2,
    'qkv_dim': 3,
    'extra_feature': False,
    'sqrt_embedding_dim': 2.0,
    'logit_clipping': 10,
    'detach_from_encoder': False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SINGLE_Decoder — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a SINGLE_Decoder output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
multi_head_attention Ran
ran; our contract draft was wrong, not the code
SwonGao/DeCoST/DeCoST/models/SINGLEModel.py
code served (permissive licence) · get_code("144218ff791ec47b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 3, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Define tensor dimensions
batch_s = 2
head_num = 3
n = 4
key_dim = 5
input_s = 6

# Construct query, key, value tensors
# q shape: (batch_s, head_num, n, key_dim)
q = torch.randn(batch_s, head_num, n, key_dim)

# k shape: (batch_s, head_num, input_s, key_dim)
k = torch.randn(batch_s, head_num, input_s, key_dim)

# v shape: (batch_s, head_num, input_s, key_dim)
v = torch.randn(batch_s, head_num, input_s, key_dim)

# rank2_ninf_mask: shape (batch_s, input_s) - broadcasts to (batch_s, 1, 1, input_s)
rank2_ninf_mask = torch.full((batch_s, input_s), float('-inf'))

# rank3_ninf_mask: shape (batch_s, n, input_s) - broadcasts to (batch_s, 1, n, input_s)
rank3_ninf_mask = torch.full((batch_s, n, input_s), float('-inf'))

# edge_bias: shape (batch_s, head_num, n, input_s)
edge_bias = torch.randn(batch_s, head_num, n, input_s)

ARGS = {
    'q': q,
    'k': k,
    'v': v,
    'rank2_ninf_mask': rank2_ninf_mask,
    'rank3_ninf_mask': rank3_ninf_mask,
    'edge_bias': edge_bias,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float — no output shape is stated: observe_signature.py could not observe it (no_result)
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
EncoderLayer Not yet run
unverified (harvested, no recorded run)
SwonGao/DeCoST/DeCoST/models/SINGLEModel.py
code served (permissive licence) · get_code("4ed91dc169a26cb2")
How it ran

no invocation recorded.

SINGLEModel Not yet run
unverified (harvested, no recorded run)
SwonGao/DeCoST/DeCoST/models/SINGLEModel.py
code served (permissive licence) · get_code("7100ae2194b3f842")
How it ran

no invocation recorded.

SINGLE_Encoder Not yet run
unverified (harvested, no recorded run)
SwonGao/DeCoST/DeCoST/models/SINGLEModel.py
code served (permissive licence) · get_code("2fc9ebcca0452e7e")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 7 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

The orienteering problem with time windows and variable profits (OPTWVP) is common in many real-world applications and involves continuous time variables. Current approaches fail to develop an efficient solver for this orienteering problem variant with discrete and continuous variables. In this paper, we propose a learningbased two-stage DEcoupled discrete-Continuous optimization with Service-timeguided Trajectory (DeCoST), which aims to effectively decouple the discrete and continuous decision variables in the OPTWVP problem, while enabling efficient and learnable coordination between them. In the first stage, a parallel decoding structure is employed to predict the path and the initial service time allocation. The second stage optimizes the service times through a linear programming (LP) formulation and provides a long-horizon learning of structure estimation. We rigorously prove the global optimality of the second-stage solution. Experiments on OPTWVP instances demonstrate that DeCoST outperforms both state-of-the-art constructive solvers and the latest meta-heuristic algorithms in terms of solution quality and computational efficiency, achieving up to 6.6x inference speedup on instances with fewer than 500 nodes. Moreover, the proposed framework is compatible with various constructive solvers and consistently enhances the solution quality for OPTWVP.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2603.06260")
get_code_for_paper("2603.06260")
have("2603.06260")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2603.06260.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2603.06260.svg)](https://syntology.ai/paper/2603.06260)

Connect an agent — have() is free.