SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · aaai_25540 · AAAI · 2023

Uniform Sequence Better: Time Interval Aware Data Augmentation for Sequential Recommendation

Hong Liu, Enneng Yang, Guibing Guo, Xingwei Wang, Xiaoxiao Xu, Qinghui Sun, Yizhou Dang, Linying Jiang

Open in the Atlas

Code that ran

10 of 10 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; 2 honoured the contract we drafted, 0 violated it and 6 have no contract verdict; 2 failures of our instrument; 0 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/aaai_25540.json · badge: /api/ran/aaai_25540.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
KingGugu/TiCoSeRec canonical (classifier label)
classifier basis: census
10 ran · 0 no recorded run
FunctionStatusWhere it lives
convert_time Ran
ran on a synthesized input
KingGugu/TiCoSeRec/data_process/json2csv_yelp.py
code served (permissive licence) · get_code("737dfde74d9ead3d")
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 float, 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 time

# The function convert_time takes a date string in format '%Y-%m-%d %H:%M:%S'
# and converts it to a timestamp using time.strptime and time.mktime

date_string = '2023-06-15 14:30:45'

ARGS = {
    'date': date_string
}
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 float, 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
convert_unique_idx Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
KingGugu/TiCoSeRec/data_process/json2csv_amazon.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from haolun-wu/jmefairness, is pointer only (licence: NONE) · get_code("8620a395f0a3a8ad")
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 haolun-wu/jmefairness, 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.

get_attention_mask Ran
ran on a synthesized input
KingGugu/TiCoSeRec/src/modules.py
code served (permissive licence) · get_code("25fc45802d67d296")
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 4, dtype family float, shape (2, 1, 4, 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)

# item_seq should be a 2D tensor (batch_size, seq_length) with integer values
# The function checks item_seq != 0, so we need some zeros and some non-zeros
item_seq = torch.tensor([
    [1, 2, 3, 0],
    [4, 5, 0, 0]
], dtype=torch.long)

bidirectional = False

ARGS = {
    "item_seq": item_seq,
    "bidirectional": bidirectional
}
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 4, dtype family float, shape (2, 1, 4, 4) — shape (2, 1, 4, 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
get_org_rank_users Ran
ran on a synthesized input
KingGugu/TiCoSeRec/data_process/csv2txt.py
code served (permissive licence) · get_code("3148e454721a6e3d")
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 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 pandas as pd

# Create a simple dataframe with a 'user' column containing some unique users
dataframe = pd.DataFrame({
    'user': ['user1', 'user2', 'user1', 'user3', 'user2'],
    'data': [10, 20, 30, 40, 50]
})

ARGS = {
    'dataframe': dataframe
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — 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 float — 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
get_var Ran
ran on a synthesized input
KingGugu/TiCoSeRec/src/data_augmentation_time.py
code served (permissive licence) · get_code("9df44c2c801091f2")
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 float, 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 numpy as np

torch.manual_seed(0)

# Analyze the function:
# - get_var takes tlist (a list or sequence of numeric values)
# - It computes variance of differences between consecutive elements
# - It supports single element lists (returns 0)
# - It needs to support len(), indexing with [], and arithmetic operations

# Create a simple list of numeric values
tlist = [1.0, 2.5, 3.1, 5.0, 4.2]

ARGS = {
    "tlist": tlist
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, 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
get_var_rank_users Ran
ran on a synthesized input
KingGugu/TiCoSeRec/data_process/csv2txt.py
code served (permissive licence) · get_code("1f1c0ccb5e0979db")
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 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 pandas as pd
import numpy as np

# Construct a minimal valid dataframe for get_var_rank_users
# The function expects columns: 'user', 'time'
# It groups by user and calculates variance of time differences

data = {
    'user': ['user1', 'user1', 'user1', 'user2', 'user2', 'user2', 'user3', 'user3'],
    'time': [1.0, 3.0, 7.0, 2.0, 4.0, 5.0, 10.0, 12.0]
}

dataframe = pd.DataFrame(data)

ARGS = {
    'dataframe': dataframe
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — 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 float — 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
k_core_check Ran
ran on a synthesized input
KingGugu/TiCoSeRec/data_process/json2csv_yelp.py
code served (permissive licence) · get_code("bda17bc987d2da5a")
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 bool, 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 pandas as pd

# Create a sample DataFrame with 'user' and 'item' columns
# The function checks if users and items meet minimum occurrence thresholds
df = pd.DataFrame({
    'user': [1, 1, 1, 2, 2, 3, 3, 3, 3],
    'item': [10, 11, 12, 10, 11, 12, 13, 14, 15]
})

# u_core: minimum number of times a user must appear
u_core = 2

# i_core: minimum number of times an item must appear
i_core = 2

ARGS = {
    'df': df,
    'u_core': u_core,
    'i_core': i_core
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, 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
kmax_pooling Ran
ran; our contract draft was wrong, not the code
KingGugu/TiCoSeRec/src/utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from RUCAIBox/CIKM2020-S3Rec, is pointer only (licence: NONE) · get_code("11e0bc26b5cfda51")
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 RUCAIBox/CIKM2020-S3Rec, 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.

nCr Ran
ran, honoured the contract we drafted
KingGugu/TiCoSeRec/src/utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from graph-team/apgl4sr, is pointer only (licence: NONE) · get_code("941f08ce28f5a2ff")
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 graph-team/apgl4sr, 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.

neg_sample Ran
ran, honoured the contract we drafted
KingGugu/TiCoSeRec/src/utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from RUCAIBox/CIKM2020-S3Rec, is pointer only (licence: NONE) · get_code("ff0f6a85cdaf16a5")
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 RUCAIBox/CIKM2020-S3Rec, 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.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 0 the same month, 3 later) and 6 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
neg_sample 7 other papers
2 dated earlier, 2 later, 3 same year, order not determinable
2202.02519 (2022-02)
Intent Contrastive Learning for Sequential Recommendation
2203.15508 (2022-03)
Improving Contrastive Learning with Model Augmentation
2310.14318 (2023-10)
Intent Contrastive Learning with Cross Subsequences for Sequential Recommendation
and 4 more not named here
kmax_pooling 6 other papers
2 dated earlier, 2 later, 2 same year, order not determinable
2202.02519 (2022-02)
Intent Contrastive Learning for Sequential Recommendation
2203.15508 (2022-03)
Improving Contrastive Learning with Model Augmentation
2310.14318 (2023-10)
Intent Contrastive Learning with Cross Subsequences for Sequential Recommendation
and 3 more not named here
nCr 5 other papers
2 dated earlier, 1 later, 2 same year, order not determinable
2202.02519 (2022-02)
Intent Contrastive Learning for Sequential Recommendation
2203.15508 (2022-03)
Improving Contrastive Learning with Model Augmentation
2310.14318 (2023-10)
Intent Contrastive Learning with Cross Subsequences for Sequential Recommendation
and 2 more not named here
convert_unique_idx 2 other papers
2 dated earlier
1205.2618 (2012-05)
BPR: Bayesian Personalized Ranking from Implicit Feedback
2205.00048 (2022-05)
Joint Multisided Exposure Fairness for Recommendation

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

Sequential recommendation is an important task to predict the next-item to access based on a sequence of interacted items. Most existing works learn user preference as the transition pattern from the previous item to the next one, ignoring the time interval between these two items. However, we observe that the time interval in a sequence may vary significantly different, and thus result in the ineffectiveness of user modeling due to the issue of preference drift. In fact, we conducted an empirical study to validate this observation, and found that a sequence with uniformly distributed time interval (denoted as uniform sequence) is more beneficial for performance improvement than that with greatly varying time interval. Therefore, we propose to augment sequence data from the perspective of time interval, which is not studied in the literature. Specifically, we design five operators (Ti-Crop, Ti-Reorder, Ti-Mask, Ti-Substitute, Ti-Insert) to transform the original non-uniform sequence to uniform sequence with the consideration of variance of time intervals. Then, we devise a control strategy to execute data augmentation on item sequences in different lengths. Finally, we implement these improvements on a state-of-the-art model CoSeRec and validate our approach on four real datasets. The experimental results show that our approach reaches significantly better performance than the other 9 competing methods. Our implementation is available: https://github.com/KingGugu/TiCoSeRec.

For agents

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

get_harvested_code_for_paper("aaai_25540")
get_code_for_paper("aaai_25540")
have("aaai_25540")

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

curl https://syntology.ai/api/ran/aaai_25540.json

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

[![Syntology run record](https://syntology.ai/api/ran/aaai_25540.svg)](https://syntology.ai/paper/aaai_25540)

Connect an agent — have() is free.