SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2603.08032 · 2026

GCGNet: Graph-Consistent Generative Network for Time Series Forecasting with Exogenous Variables

Bin Yang, Chenjuan Guo, Yuhan Zhu, Xingjian Wu, Xiangfei Qiu, Zhengyu Li, Jilin Hu, � 𝒀𝒀

arXiv · PDF · Open in the Atlas

Code that ran

10 of 13 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (8 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 9 have no contract verdict; 1 failure of our instrument; 3 no recorded run), measured 2026-11-21; 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/2603.08032.json · badge: /api/ran/2603.08032.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
decisionintelligence/GCGNet community
Papers with Code archive
10 ran · 3 no recorded run
FunctionStatusWhere it lives
FlattenHead Ran
ran on a synthesized input · constructed an object; no result was computed
decisionintelligence/GCGNet/ts_benchmark/baselines/GCGNet/models/gcgnet_model.py
code served (permissive licence) · get_code("68b22dc86911e508")
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, FlattenHead: 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 parameters for FlattenHead
n_vars = 5  # number of variables
endo_num = 3  # number of endogenous variables
nf = 15  # number of features (input to MLP)
d_ff = 32  # hidden dimension of MLP
target_window = 8  # output window size
head_dropout = 0.1  # dropout rate

ARGS = {
    "n_vars": n_vars,
    "endo_num": endo_num,
    "nf": nf,
    "d_ff": d_ff,
    "target_window": target_window,
    "head_dropout": head_dropout,
}
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 FlattenHead — 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 FlattenHead 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
GCN Ran
ran on a synthesized input
decisionintelligence/GCGNet/ts_benchmark/baselines/GCGNet/models/gcgnet_model.py
code served (permissive licence) · get_code("39a5436e3ee292c2")
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 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 numpy as np
from scipy.sparse import csr_matrix

INIT_ARGS = {
    'd_model': 4,
    'n_heads': 2
}

# Create small stand-in tensors for adj and x
B, L, D = 2, 3, 4  # batch=2, sequence_length=3, feature_dim=4
adj = torch.rand(B, INIT_ARGS['n_heads'], L, L)  # (B, n_heads, L, L)
x = torch.rand(B, L, D)  # (B, L, D)

ARGS = {
    'adj': adj,
    'x': x
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
numpy, scipy, 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 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)
GCNStack Ran
ran on a synthesized input · constructed an object; no result was computed
decisionintelligence/GCGNet/ts_benchmark/baselines/GCGNet/models/gcgnet_model.py
code served (permissive licence) · get_code("df1d7372826b2b11")
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, GCNStack: 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 torch import nn

torch.manual_seed(0)

# Define parameters for GCNStack
d_model = 4
n_heads = 2
e_layers = 2
dropout = 0.1

# Create adjacency matrix: (batch, seq_len, seq_len)
batch_size = 2
seq_len = 3
adj = torch.randn(batch_size, seq_len, seq_len)
adj = torch.softmax(adj, dim=-1)  # normalize to be valid adjacency

# Create input tensor: (batch, seq_len, d_model)
x = torch.randn(batch_size, seq_len, d_model)

ARGS = {
    'd_model': d_model,
    'n_heads': n_heads,
    'e_layers': e_layers,
    'dropout': dropout,
}
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 GCNStack — 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 GCNStack 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
GraphDiscriminator Ran
ran on a synthesized input · constructed an object; no result was computed
decisionintelligence/GCGNet/ts_benchmark/baselines/GCGNet/models/gcgnet_model.py
code served (permissive licence) · get_code("ab0710398841c1db")
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, GraphDiscriminator: 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 = {
    "nodes_num": 5,
    "d_model": 8,
    "d_ff": 16,
    "n_heads": 2,
    "rank": 3,
}
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 GraphDiscriminator — 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 GraphDiscriminator 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
GraphLearner Ran
ran on a synthesized input · constructed an object; no result was computed
decisionintelligence/GCGNet/ts_benchmark/baselines/GCGNet/models/gcgnet_model.py
code served (permissive licence) · get_code("6a98b754f69d8aad")
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, GraphLearner: 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 torch import nn
import torch.nn.functional as F

class GraphLearner(nn.Module):

    def __init__(self, d_model, n_heads):
        super().__init__()
        self.n_heads = n_heads
        self.proj_1 = nn.Linear(d_model, d_model)
        self.proj_2 = nn.Linear(d_model, d_model)

    def forward(self, x):
        B, P_NUM, _ = x.shape
        proj_1_x = self.proj_1(x).view(B, P_NUM, self.n_heads, -1).permute(0, 2, 1, 3)
        proj_2_x = self.proj_2(x).view(B, P_NUM, self.n_heads, -1).permute(0, 2, 1, 3)
        adj = F.gelu(torch.einsum('bhid,bhjd->bhij', proj_1_x, proj_2_x)).contiguous()
        adj = 0.5 * (adj + adj.transpose(-1, -2))
        return adj

torch.manual_seed(0)

d_model = 8
n_heads = 2

ARGS = {
    "d_model": d_model,
    "n_heads": n_heads
}
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 GraphLearner — 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 GraphLearner 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
GraphVAE Ran
ran on a synthesized input · constructed an object; no result was computed
decisionintelligence/GCGNet/ts_benchmark/baselines/GCGNet/models/gcgnet_model.py
code served (permissive licence) · get_code("6a84c055bb52f17b")
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, GraphVAE: 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 torch import nn

torch.manual_seed(0)

ARGS = {
    "d_model": 8,
    "d_ff": 16,
    "max_num_nodes": 4,
}
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 GraphVAE — 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 GraphVAE 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
PatchEmbedding Ran
ran on a synthesized input · constructed an object; no result was computed
decisionintelligence/GCGNet/ts_benchmark/baselines/GCGNet/models/gcgnet_model.py
code served (permissive licence) · get_code("ce9179b982e2a774")
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, PatchEmbedding: 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
import math

torch.manual_seed(0)

ARGS = {
    'd_model': 8,
    'patch_len': 4,
    'stride': 2,
    'dropout': 0.1
}
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 PatchEmbedding — 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 PatchEmbedding 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
PositionalEmbedding Ran
ran on a synthesized input · constructed an object; no result was computed
decisionintelligence/GCGNet/ts_benchmark/baselines/GCGNet/models/gcgnet_model.py
code served (permissive licence) · get_code("a252f6205f6f5eae")
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, PositionalEmbedding: 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 math
import torch
from torch import nn

torch.manual_seed(0)

class PositionalEmbedding(nn.Module):

    def __init__(self, d_model, max_len=5000):
        super(PositionalEmbedding, self).__init__()
        pe = torch.zeros(max_len, d_model).float()
        pe.require_grad = False
        position = torch.arange(0, max_len).float().unsqueeze(1)
        div_term = (torch.arange(0, d_model, 2).float() * -(math.log(10000.0) / d_model)).exp()
        pe[:, 0::2] = torch.sin(position * div_term)
        pe[:, 1::2] = torch.cos(position * div_term)
        pe = pe.unsqueeze(0)
        self.register_buffer('pe', pe)

    def forward(self, x):
        return self.pe[:, :x.size(1)]

# Construct test inputs for PositionalEmbedding
d_model = 4
max_len = 100

ARGS = {
    "d_model": d_model,
    "max_len": max_len
}
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 PositionalEmbedding — 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 PositionalEmbedding 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
VAE Ran
ran on a synthesized input · constructed an object; no result was computed
decisionintelligence/GCGNet/ts_benchmark/baselines/GCGNet/models/gcgnet_model.py
code served (permissive licence) · get_code("bb440cf19c0f42eb")
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, VAE: 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 torch import nn

torch.manual_seed(0)

ARGS = {
    "input_len": 8,
    "output_len": 6,
    "d_model": 16,
    "d_ff": 32,
    "latent_dim": 8,
}
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 VAE — 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 VAE 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
mask_topk Ran
ran; our contract draft was wrong, not the code
decisionintelligence/GCGNet/ts_benchmark/baselines/GCGNet/models/gcgnet_model.py
code served (permissive licence) · get_code("6d24b7638890ca20")
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 2, dtype family float, shape (2, 4) 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)

# Create a small input tensor with shape (batch=2, dim=4)
x = torch.randn(2, 4)

# Define ARGS dict with all parameters
ARGS = {
    "x": x,
    "alpha": 0.5,
    "largest": 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 Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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
GCGNetModel Not yet run
unverified (harvested, no recorded run)
decisionintelligence/GCGNet/ts_benchmark/baselines/GCGNet/models/gcgnet_model.py
code served (permissive licence) · get_code("6906578b0c16cc33")
How it ran

no invocation recorded.

Sparsifier Not yet run
unverified (harvested, no recorded run)
decisionintelligence/GCGNet/ts_benchmark/baselines/GCGNet/models/gcgnet_model.py
code served (permissive licence) · get_code("0a9dd3176596d414")
How it ran

no invocation recorded.

mask_moe Not yet run
unverified (harvested, no recorded run)
decisionintelligence/GCGNet/ts_benchmark/baselines/GCGNet/models/gcgnet_model.py
code served (permissive licence) · get_code("c962a0d3411c2bf5")
How it ran

no invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 1 later) and 11 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
PositionalEmbedding
brick of this paper's assembly: helper
19 other papers
18 dated earlier, 1 later · 16 only as an assembly brick
2012.07436 (2020-12)
Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting
2106.13008 (2021-06)
Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting
2201.12740 (2022-01)
FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecasting
and 16 more not named here
mask_topk
brick of this paper's assembly: helper
3 other papers
3 dated earlier · 3 only as an assembly brick
2509.23597 (2025-09)
Characteristic Root Analysis and Regularization for Linear Time Series Forecasting
2601.23114 (2026-01)
To See Far, Look Close: Evolutionary Forecasting for Long-term Time Series
2602.00297 (2026-02)
From Observations to States: Latent Time Series Forecasting

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 2026-11-21.

Repositories linked to this paper

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

Abstract

Exogenous variables offer valuable supplementary information for predicting future endogenous variables. Forecasting with exogenous variables needs to consider both past-to-future dependencies (i.e., temporal correlations) and the influence of exogenous variables on endogenous variables (i.e., channel correlations). This is pivotal when future exogenous variables are available, because they may directly affect the future endogenous variables. Many methods have been proposed for time series forecasting with exogenous variables, focusing on modeling temporal and channel correlations. However, most of them use a two-step strategy, modeling temporal and channel correlations separately, which limits their ability to capture joint correlations across time and channels. Furthermore, in realworld scenarios, time series are frequently affected by various forms of noises, underscoring the critical importance of robustness in such correlations modeling. To address these limitations, we propose GCGNet, a Graph-Consistent Generative Network for time series forecasting with exogenous variables. Specifically, GCGNet first employs a Variational Generator to produce coarse predictions. A Graph Structure Aligner then further guides it by evaluating the consistency between the generated and true correlations, where the correlations are represented as graphs, and are robust to noises. Finally, a Graph Refiner is proposed to refine the predictions to prevent degeneration and improve accuracy. Extensive experiments on 12 real-world datasets demonstrate that GCGNet outperforms state-of-the-art baselines. We make our code and datasets available at https://github.com/decisionintelligence/GCGNet.

For agents

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

get_harvested_code_for_paper("2603.08032")
get_code_for_paper("2603.08032")
have("2603.08032")

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

curl https://syntology.ai/api/ran/2603.08032.json

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

[![Syntology run record](https://syntology.ai/api/ran/2603.08032.svg)](https://syntology.ai/paper/2603.08032)

Connect an agent — have() is free.