SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2305.19587 · ICML · 2023

Towards Omni-generalizable Neural Methods for Vehicle Routing Problems

Jie Zhang, Zhiguang Cao, Wen Song, Jianan Zhou, Yaoxin Wu

arXiv · PDF · Open in the Atlas

Code that ran

13 of 22 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; 3 honoured the contract we drafted, 0 violated it and 5 have no contract verdict; 5 failures of our instrument; 9 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; 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/2305.19587.json · badge: /api/ran/2305.19587.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
royalskye/omni-vrp canonical (classifier label)
classifier basis: classifier_2025
10 ran · 6 no recorded run
RoyalSkye/Omni-VRP canonical (classifier label)
classifier basis: census
3 ran · 3 no recorded run
FunctionStatusWhere it lives
EncoderLayer Ran
ran on a synthesized input · constructed an object; no result was computed
royalskye/omni-vrp/POMO/CVRP/CVRPTrainer_meta.py
code served (permissive licence) · get_code("27459636e9d1f27b")
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, EncoderLayer: 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 = {
    'embedding_dim': 16,
    'head_num': 4,
    'qkv_dim': 4,
    'ff_hidden_dim': 32,
    'norm': 'batch',
}
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 EncoderLayer — 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 EncoderLayer 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
FeedForward Ran
ran on a synthesized input
royalskye/omni-vrp/POMO/CVRP/CVRPTrainer_meta.py
code served (permissive licence) · get_code("1906754f665b381b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 2, 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

INIT_ARGS = {
    'embedding_dim': 4,
    'ff_hidden_dim': 8,
}

ARGS = {
    'input1': torch.randn(2, 4),
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 2, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
Reset_State Ran
ran on a synthesized input · constructed an object; no result was computed
royalskye/omni-vrp/POMO/CVRP/CVRPTrainer_meta.py
code served (permissive licence) · get_code("deed3a52d6df807a")
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, Reset_State: 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
from dataclasses import dataclass

torch.manual_seed(0)

@dataclass
class Reset_State:
    depot_xy: torch.Tensor = None
    node_xy: torch.Tensor = None
    node_demand: torch.Tensor = None

# Create small tensors with distinct sizes
depot_xy = torch.randn(2, 2)  # batch=2, coordinates=2
node_xy = torch.randn(2, 5, 2)  # batch=2, num_nodes=5, coordinates=2
node_demand = torch.randn(2, 5)  # batch=2, num_nodes=5

ARGS = {
    "depot_xy": depot_xy,
    "node_xy": node_xy,
    "node_demand": node_demand,
}
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 Reset_State — 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 Reset_State 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
Step_State Ran
ran on a synthesized input · constructed an object; no result was computed
royalskye/omni-vrp/POMO/CVRP/CVRPTrainer_meta.py
code served (permissive licence) · get_code("57c0114e6f1f1079")
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, Step_State: 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
from dataclasses import dataclass

torch.manual_seed(0)

@dataclass
class Step_State:
    BATCH_IDX: torch.Tensor = None
    POMO_IDX: torch.Tensor = None
    selected_count: int = None
    load: torch.Tensor = None
    current_node: torch.Tensor = None
    ninf_mask: torch.Tensor = None
    finished: torch.Tensor = None

ARGS = {
    "BATCH_IDX": torch.tensor([0, 1], dtype=torch.long),
    "POMO_IDX": torch.tensor([0, 1, 2, 3], dtype=torch.long),
    "selected_count": 5,
    "load": torch.randn(2, 4),
    "current_node": torch.tensor([[1, 2], [3, 4]], dtype=torch.long),
    "ninf_mask": torch.zeros(2, 4, dtype=torch.bool),
    "finished": torch.tensor([False, False], dtype=torch.bool),
}
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 Step_State — 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 Step_State 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
augment_xy_data_by_8_fold Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
royalskye/omni-vrp/POMO/CVRP/CVRPTrainer_meta.py
code served (permissive licence) · get_code("ae7acd105c3e55ab")
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, shape (16, 4, 2) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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

torch.manual_seed(0)

# The function expects 'problems' to be a tensor with shape (batch, seq_len, 2)
# It extracts x and y from indices 0 and 1 of the last dimension
# Then performs various augmentations and concatenations

problems = torch.rand(2, 4, 2)

ARGS = {
    "problems": problems
}
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, shape (16, 4, 2) — shape (16, 4, 2) as a later run recorded it
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
augment_xy_data_by_8_fold Ran
ran on a synthesized input
RoyalSkye/Omni-VRP/EAS/source/utilities.py
code served (permissive licence) · get_code("a36b58e1d6f73f99")
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, shape (16, 4, 2) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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

torch.manual_seed(0)

# The function expects xy_data to be a 3D tensor with shape (batch, sequence, features)
# where features contains x and y coordinates that should be in range [0, 1]
# Shape: (batch_size=2, sequence_length=4, features=2)
xy_data = torch.rand(2, 4, 2)

ARGS = {
    "xy_data": xy_data
}
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, shape (16, 4, 2) — shape (16, 4, 2) as a later run recorded it
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
check_extension Ran
ran; our contract draft was wrong, not the code
royalskye/omni-vrp/POMO/CVRP/CVRPTrainer_meta.py
code served (permissive licence) · get_code("01df1926cf07ac89")
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 str. 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 os

def check_extension(filename):
    if os.path.splitext(filename)[1] != '.pkl':
        return filename + '.pkl'
    return filename

ARGS = {
    'filename': 'myfile.txt'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
generate_task_set Ran
ran, honoured the contract we drafted
royalskye/omni-vrp/POMO/CVRP/CVRPTrainer_meta.py
code served (permissive licence) · get_code("ee885e080743b365")
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 list, ndim 2, dtype family int; 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

torch.manual_seed(0)

ARGS = {
    'meta_params': {
        'data_type': 'distribution'
    }
}
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 list, ndim 2, dtype family int — a list output has no array shape of its own; none is stated
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
generate_tsp_dist Ran
ran, honoured the contract we drafted
royalskye/omni-vrp/POMO/CVRP/CVRPTrainer_meta.py
code served (permissive licence) · get_code("75f79d4f7fbc6f14")
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 ndarray, ndim 3, dtype family float, shape (2, 5, 2) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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 numpy as np

ARGS = {
    "n_samples": 2,
    "n_nodes": 5,
    "distribution": "uniform_rectangle"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family float, shape (2, 5, 2) — shape (2, 5, 2) as a later run recorded it
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
load_dataset Ran
ran, honoured the contract we drafted
royalskye/omni-vrp/POMO/CVRP/CVRPTrainer_meta.py
code served (permissive licence) · get_code("04edb896dec75a82")
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 list, ndim 1, dtype family int; 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 os
import pickle
import tempfile

# Create a temporary pickle file for testing
temp_dir = tempfile.mkdtemp()
test_data = [1, 2, 3, 4, 5]
test_file_path = os.path.join(temp_dir, "test_data")

with open(test_file_path + ".pkl", "wb") as f:
    pickle.dump(test_data, f)

ARGS = {
    "filename": test_file_path,
    "disable_print": False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family int — a list output has no array shape of its own; none is stated
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
multi_head_attention Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
RoyalSkye/Omni-VRP/POMO/CVRP/CVRPModel.py
pointer only (licence: MIT); the stored copy get_code serves, from SobihanSurendran/LGS, is pointer only (licence: NONE) · get_code("6dc103e776ff9eed")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from SobihanSurendran/LGS, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

reshape_by_heads Ran
ran; our contract draft was wrong, not the code
RoyalSkye/Omni-VRP/POMO/CVRP/CVRPModel.py
pointer only (licence: MIT); the stored copy get_code serves, from yihang0410/RADAR, is pointer only (licence: NONE) · get_code("459fe521069329c6")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from yihang0410/RADAR, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

run_all_in_pool Ran
ran; our contract draft was wrong, not the code
royalskye/omni-vrp/POMO/CVRP/CVRPTrainer_meta.py
code served (permissive licence) · get_code("4a0fdf421df64174")
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 tuple. 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 os
from multiprocessing import Pool
from multiprocessing.dummy import Pool as ThreadPool

def dummy_func(args):
    """Dummy function to process items in pool"""
    directory, idx, *problem = args
    # Simulate successful processing
    return True

class DummyOpts:
    def __init__(self):
        self.cpus = None
        self.n = None
        self.offset = None
        self.progress_bar_mininterval = 0.1

class DummyDataset:
    def __init__(self, data):
        self.data = data
    
    def __getitem__(self, idx):
        if isinstance(idx, slice):
            return self.data[idx]
        return self.data[idx]
    
    def __len__(self):
        return len(self.data)

# Create a minimal dataset: list of tuples that can be unpacked
dataset_items = [
    (1, 2, 3),
    (4, 5, 6),
    (7, 8, 9),
    (10, 11, 12),
]
dataset = DummyDataset(dataset_items)

opts = DummyOpts()
opts.cpus = 1
opts.n = None
opts.offset = 0
opts.progress_bar_mininterval = 0.1

ARGS = {
    'func': dummy_func,
    'directory': '/tmp/test_output',
    'dataset': dataset,
    'opts': opts,
    'use_multiprocessing': False,
    'disable_tqdm': True,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is 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
Add_And_Normalization_Module Not yet run
unverified (harvested, no recorded run)
royalskye/omni-vrp/POMO/CVRP/CVRPTrainer_meta.py
code served (permissive licence) · get_code("f3649a7457360ebc")
How it ran

no invocation recorded.

CVRPTrainer Not yet run
unverified (harvested, no recorded run)
royalskye/omni-vrp/POMO/CVRP/CVRPTrainer_meta.py
code served (permissive licence) · get_code("eaff33fa84aabf33")
How it ran

no invocation recorded.

CVRP_Decoder Not yet run
unverified (harvested, no recorded run)
royalskye/omni-vrp/POMO/CVRP/CVRPTrainer_meta.py
code served (permissive licence) · get_code("4bd04c65381e358c")
How it ran

no invocation recorded.

CVRP_Encoder Not yet run
unverified (harvested, no recorded run)
royalskye/omni-vrp/POMO/CVRP/CVRPTrainer_meta.py
code served (permissive licence) · get_code("bc46074bd008d171")
How it ran

no invocation recorded.

generate_gaussian_mixture_tsp Not yet run
unverified (harvested, no recorded run)
royalskye/omni-vrp/POMO/CVRP/CVRPTrainer_meta.py
code served (permissive licence) · get_code("7fbb43e1922d836c")
How it ran

no invocation recorded.

get_random_problems Not yet run
unverified (harvested, no recorded run)
royalskye/omni-vrp/POMO/CVRP/CVRPTrainer_meta.py
code served (permissive licence) · get_code("1e4f1088fddcd277")
How it ran

no invocation recorded.

get_random_problems Not yet run
unverified (harvested, no recorded run)
RoyalSkye/Omni-VRP/POMO/ProblemDef.py
code served (permissive licence) · get_code("f2e0d7557ce1218c")
How it ran

no invocation recorded.

run_eas_emb Not yet run
unverified (harvested, no recorded run)
RoyalSkye/Omni-VRP/EAS/source/eas_emb.py
code served (permissive licence) · get_code("6100c5076837bdee")
How it ran

no invocation recorded.

run_eas_tab Not yet run
unverified (harvested, no recorded run)
RoyalSkye/Omni-VRP/EAS/source/eas_tab.py
code served (permissive licence) · get_code("f0530713ee5b83cb")
How it ran

no invocation recorded.

Also present in other papers' code

Of 22 samples lifted from this paper's linked repositories, 22 carry a function identity and 0 do not. Of the 21 distinct functions with an identity, 4 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 4 later) and 17 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.

FunctionAlso present inEarliest-datedOthers named
reshape_by_heads
brick of this paper's assembly: helper
28 other papers
6 dated earlier, 21 later, 1 undated · 15 only as an assembly brick
2010.16011 (2020-10)
POMO: Policy Optimization with Multiple Optima for Reinforcement Learning
2106.11113 (2021-06)
Matrix Encoding Networks for Neural Combinatorial Optimization
2203.15386 (2022-03)
Pareto Set Learning for Neural Multi-objective Combinatorial Optimization
and 25 more not named here
multi_head_attention
brick of this paper's assembly: helper
13 other papers
4 dated earlier, 8 later, 1 undated · 6 only as an assembly brick
2010.16011 (2020-10)
POMO: Policy Optimization with Multiple Optima for Reinforcement Learning
2203.15386 (2022-03)
Pareto Set Learning for Neural Multi-objective Combinatorial Optimization
2205.13209 (2022-05)
Sym-NCO: Leveraging Symmetricity for Neural Combinatorial Optimization
and 10 more not named here
check_extension
brick of this paper's assembly: helper
7 other papers
2 dated earlier, 5 later
1803.08475 (2018-03)
Attention, Learn to Solve Routing Problems!
2205.13225 (2022-05)
DevFormer: A Symmetric Transformer for Context-Aware Device Placement
2308.14104 (2023-08)
Towards Generalizable Neural Solvers for Vehicle Routing Problems via Ensemble with Transferrable Local Policy
and 4 more not named here
augment_xy_data_by_8_fold
brick of this paper's assembly: helper
3 other papers
3 later
2308.14104 (2023-08)
Towards Generalizable Neural Solvers for Vehicle Routing Problems via Ensemble with Transferrable Local Policy
ijcai2024_0764 (2024 (year only))
Towards Generalizable Neural Solvers for Vehicle Routing Problems via Ensemble with Transferrable Local Policy
2509.23413 (2025-09)
URS: A Unified Neural Routing Solver for Cross-Problem Zero-Shot Generalization

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

Learning heuristics for vehicle routing problems (VRPs) has gained much attention due to the less reliance on hand-crafted rules. However, existing methods are typically trained and tested on the same task with a fixed size and distribution (of nodes), and hence suffer from limited generalization performance. This paper studies a challenging yet realistic setting, which considers generalization across both size and distribution in VRPs. We propose a generic meta-learning framework, which enables effective training of an initialized model with the capability of fast adaptation to new tasks during inference. We further develop a simple yet efficient approximation method to reduce the training overhead. Extensive experiments on both synthetic and benchmark instances of the traveling salesman problem (TSP) and capacitated vehicle routing problem (CVRP) demonstrate the effectiveness of our method. The code is available at: https: //github.com/RoyalSkye/Omni-VRP.

For agents

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

get_harvested_code_for_paper("2305.19587")
get_code_for_paper("2305.19587")
have("2305.19587")

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

curl https://syntology.ai/api/ran/2305.19587.json

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

[![Syntology run record](https://syntology.ai/api/ran/2305.19587.svg)](https://syntology.ai/paper/2305.19587)

Connect an agent — have() is free.