SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2203.07782 · ACL · 2022

Complex Evolutional Pattern Learning for Temporal Knowledge Graph Reasoning

Wei Li, Xueqi Cheng, Yajuan Lyu, Zixuan Li, Long Bai, Jiafeng Guo, Weihua Peng, Yong Zhu, Xiaolong Jin, Saiping Guan

arXiv · PDF · Open in the Atlas

Code that ran

4 of 4 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 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; 0 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/2203.07782.json · badge: /api/ran/2203.07782.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
Lee-zix/CEN canonical (classifier label)
classifier basis not recorded on the link
4 ran · 0 no recorded run
FunctionStatusWhere it lives
sort_and_rank Ran
ran; our contract draft was wrong, not the code
Lee-zix/CEN/rgcn/utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from yecchen/gdelt-complexevent, is pointer only (licence: NONE) · get_code("f5e395388b1f6fb5")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from yecchen/gdelt-complexevent, 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.

sort_and_rank_filter Ran
ran on a synthesized input
Lee-zix/CEN/rgcn/utils.py
code served (permissive licence) · get_code("23f3fb98873f1cfa")
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 1, dtype family int, shape (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)

# batch_a: tensor of indices into all_ans (first dimension)
batch_a = torch.tensor([0, 1])

# batch_r: tensor of indices into all_ans (second dimension)
batch_r = torch.tensor([0, 1])

# score: 2D tensor of scores, shape (batch_size, num_answers)
# batch_size=2, num_answers=4
score = torch.randn(2, 4)

# target: 1D tensor of target answer indices for each sample in batch
target = torch.tensor([1, 2])

# all_ans: 2D structure that maps (batch_a[i], batch_r[i]) to a list of multi-answer indices
# all_ans[batch_a[i]][batch_r[i]] should return a list/iterable
all_ans = [
    {0: [0, 2], 1: [1, 3]},  # all_ans[0]
    {0: [0, 1], 1: [2, 3]}   # all_ans[1]
]

ARGS = {
    "batch_a": batch_a,
    "batch_r": batch_r,
    "score": score,
    "target": target,
    "all_ans": all_ans
}
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 1, dtype family int, shape (2,) — shape (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
sort_and_rank_time_filter Ran
ran on a synthesized input
Lee-zix/CEN/rgcn/utils.py
code served (permissive licence) · get_code("34b57ed02fdcb73e")
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 1, dtype family int, shape (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)

# Construct tensors for the function
batch_size = 2
num_scores = 4
embedding_dim = 3

# batch_a: anchor batch (used in loop but not directly indexed, just for len())
batch_a = torch.randn(batch_size, embedding_dim)

# batch_r: reference batch (used in loop but not directly indexed, just for len())
batch_r = torch.randn(batch_size, embedding_dim)

# score: tensor that will be sorted, shape (batch_size, num_scores)
score = torch.randn(batch_size, num_scores)

# target: tensor that will be compared with sorted indices, shape (batch_size,)
# Values should be valid indices into the score dimension
target = torch.tensor([1, 2])

# total_triplets: integer representing total number of triplets
total_triplets = 10

ARGS = {
    'batch_a': batch_a,
    'batch_r': batch_r,
    'score': score,
    'target': target,
    'total_triplets': total_triplets
}
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 1, dtype family int, shape (2,) — shape (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
temporal_regularization Ran
ran on a synthesized input
Lee-zix/CEN/src/ensemble.py
code served (permissive licence) · get_code("e58ee2f17a1b8e52")
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 0, dtype family float, shape () 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
import torch.nn as nn

torch.manual_seed(0)

# The function expects two iterables of parameters (likely from nn.Module.parameters())
# It iterates through them with zip, computes element-wise differences, and calculates L2 norms

# Create two simple models with different parameter values
model1 = nn.Linear(4, 3)
model2 = nn.Linear(4, 3)

# Get parameter lists from both models
params1 = list(model1.parameters())
params2 = list(model2.parameters())

# Modify params2 slightly so they're different from params1
for p in params2:
    p.data += 0.1

ARGS = {
    "params1": params1,
    "params2": params2,
}
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 0, dtype family float, shape () — shape () 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

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 3 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
sort_and_rank 2 other papers
2 later
2311.10112 (2023-11)
zrLLM: Zero-Shot Relational Learning on Temporal Knowledge Graphs with Large Language Models
2312.01052 (2023-12)
SCTc-TE: A Comprehensive Formulation and Benchmark for Temporal Event 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 2026-11-22.

Repositories linked to this paper

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

Abstract

A Temporal Knowledge Graph (TKG) is a sequence of KGs corresponding to different timestamps. TKG reasoning aims to predict potential facts in the future given the historical KG sequences. One key of this task is to mine and understand evolutional patterns of facts from these sequences. The evolutional patterns are complex in two aspects, lengthdiversity and time-variability. Existing models for TKG reasoning focus on modeling fact sequences of a fixed length, which cannot discover complex evolutional patterns that vary in length. Furthermore, these models are all trained offline, which cannot well adapt to the changes of evolutional patterns from then on. Thus, we propose a new model, called Complex Evolutional Network (CEN), which uses a length-aware Convolutional Neural Network (CNN) to handle evolutional patterns of different lengths via an easy-to-difficult curriculum learning strategy. Besides, we propose to learn the model under the online setting so that it can adapt to the changes of evolutional patterns over time. Extensive experiments demonstrate that CEN obtains substantial performance improvement under both the traditional offline and the proposed online settings.

For agents

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

get_harvested_code_for_paper("2203.07782")
get_code_for_paper("2203.07782")
have("2203.07782")

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

curl https://syntology.ai/api/ran/2203.07782.json

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

[![Syntology run record](https://syntology.ai/api/ran/2203.07782.svg)](https://syntology.ai/paper/2203.07782)

Connect an agent — have() is free.