SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2308.01737 · 2023

MAP: A Model-agnostic Pretraining Framework for Click-through Rate Prediction

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 0 have no contract verdict; 4 failures 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/2308.01737.json · badge: /api/ran/2308.01737.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
chiangel/map-code canonical (classifier label)
classifier basis: census
4 ran · 0 no recorded run
FunctionStatusWhere it lives
get_act Ran
ran; our contract draft was wrong, not the code
chiangel/map-code/code/layers.py
code served (permissive licence) · get_code("3778f717815c8a62")
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 ReLU. 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 math

torch.manual_seed(0)

class LEU(nn.Module):
    __constants__ = ['alpha', 'inplace']
    alpha: float
    inplace: bool

    def __init__(self, alpha: float=1.0):
        super(LEU, self).__init__()
        self.alpha = alpha

    def forward(self, input: torch.Tensor) -> torch.Tensor:
        output = torch.empty_like(input)
        output[input > 0] = self.alpha * torch.log(input[input > 0] + 1)
        output[input <= 0] = self.alpha * (torch.exp(input[input <= 0]) - 1)
        return output

class NoneAct(nn.Module):

    def forward(self, x):
        return x

class GELU(nn.Module):

    def forward(self, x):
        return x * 0.5 * (1.0 + torch.erf(x / math.sqrt(2.0)))

class GELU_new(nn.Module):

    def forward(self, x):
        return 0.5 * x * (1 + torch.tanh(math.sqrt(2 / math.pi) * (x + 0.044715 * torch.pow(x, 3))))

class Swish(nn.Module):

    def forward(self, x):
        return x * torch.sigmoid(x)

class Mish(nn.Module):

    def forward(self, x):
        return x * torch.tanh(nn.functional.softplus(x))

def get_act(act_func):
    if isinstance(act_func, str):
        if act_func.lower() == 'relu':
            return nn.ReLU(inplace=True)
        elif act_func.lower() == 'tanh':
            return nn.Tanh()
        elif act_func.lower() == 'sigmoid':
            return nn.Sigmoid()
        elif act_func.lower() == 'none':
            return NoneAct()
        elif act_func.lower() == 'elu':
            return nn.ELU()
        elif act_func.lower() == 'leu':
            return LEU()
        elif act_func.lower() == 'gelu':
            return GELU()
        elif act_func.lower() == 'gelu_new':
            return GELU_new()
        elif act_func.lower() == 'swish':
            return Swish()
        elif act_func.lower() == 'mish':
            return Mish()
        else:
            raise NotImplementedError
    else:
        return act_func

ARGS = {
    'act_func': 'relu'
}
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 ReLU — a ReLU output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ReLU 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
stat Ran
ran; our contract draft was wrong, not the code
chiangel/map-code/data_preprocess/proc_avazu.py
code served (permissive licence) · get_code("e61d2a74d4f53018")
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 numpy as np
import torch

# Test case 1: numeric data (array-like with min, max, mean, std methods)
numeric_data = np.array([1, 1, 1, 2, 2, 3, 4, 5, 5, 5, 5, -1, -1])

# Test case 2: string data (list of strings)
string_data = ['cat', 'cat', 'dog', 'bird', 'cat', 'dog', '-']

ARGS = {
    'data': numeric_data
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch, 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
trans_cat_feat Ran
ran; our contract draft was wrong, not the code
chiangel/map-code/data_preprocess/proc_criteo.py
code served (permissive licence) · get_code("6196722205ead0c1")
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 Series. 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 pandas as pd
import numpy as np
import math

# Create a small dataframe with some numeric and string columns
df = pd.DataFrame({
    'col_A': [1.0, 2.0, 3.0, np.nan, 5.0],
    'col_B': ['a', 'b', 'c', 'd', 'e']
})

# Define arguments for trans_cat_feat
ARGS = {
    'df': df,
    'col_name': 'col_A'
}
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 Series — a Series output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Series output the recorded ndim 1 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
trans_num_feat Ran
ran; our contract draft was wrong, not the code
chiangel/map-code/data_preprocess/proc_criteo.py
code served (permissive licence) · get_code("848000fb0eb95566")
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 Series. 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 numpy as np
import pandas as pd
import math

# Create a simple dataframe with numeric and edge case values
df = pd.DataFrame({
    'test_col': [1, 2, 5, 10, 100, np.nan, 3.5, 50],
    'other_col': [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8]
})

ARGS = {
    'df': df,
    'col_name': 'test_col'
}
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 Series — a Series output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Series output the recorded ndim 1 and dtype family int 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

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, 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 4 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

With the widespread application of personalized online services, click-through rate (CTR) prediction has received more and more attention and research. The most prominent features of CTR prediction are its multi-field categorical data format, and vast and daily-growing data volume. The large capacity of neural models helps digest such massive amounts of data under the supervised learning paradigm, yet they fail to utilize the substantial data to its full potential, since the 1-bit click signal is not sufficient to guide the model to learn capable representations of features and instances. The self-supervised learning paradigm provides a more promising pretrain-finetune solution to better exploit the large amount of user click logs, and learn more generalized and effective representations. However, self-supervised learning for CTR prediction is still an open question, since current works on this line are only preliminary and rudimentary. To this end, we propose a Model-agnostic pretraining (MAP) framework that applies feature corruption and recovery on multi-field categorical data, and more specifically, we derive two practical algorithms: masked feature prediction (MFP) and replaced feature detection (RFD). MFP digs into feature interactions within each instance through masking and predicting a small portion of input features, and introduces noise contrastive estimation (NCE) to handle large feature spaces. RFD further turns MFP into a binary classification mode through replacing and detecting changes in input features, making it even simpler and more effective for CTR pretraining. Our extensive experiments on two real-world large-scale datasets (i.e., Avazu, Criteo) demonstrate the advantages of these two methods on several strong backbones (e.g., DCNv2, DeepFM), and achieve new state-of-the-art performance in terms of both effectiveness and efficiency for CTR prediction.

For agents

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

get_harvested_code_for_paper("2308.01737")
get_code_for_paper("2308.01737")
have("2308.01737")

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

curl https://syntology.ai/api/ran/2308.01737.json

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

[![Syntology run record](https://syntology.ai/api/ran/2308.01737.svg)](https://syntology.ai/paper/2308.01737)

Connect an agent — have() is free.