SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2310.14047 · EMNLP · 2023

MeaeQ: Mount Model Extraction Attacks with Efficient Queries

Wei Zhou, Kun Li, Minxuan Lv, Chengwei Dai

arXiv · PDF · Open in the Atlas

Code that ran

5 of 7 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 0 have no contract verdict; 3 failures of our instrument; 2 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/2310.14047.json · badge: /api/ran/2310.14047.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
C-W-D/MeaeQ canonical (classifier label)
classifier basis: census
5 ran · 2 no recorded run
FunctionStatusWhere it lives
bert_config Ran
ran; our contract draft was wrong, not the code
C-W-D/MeaeQ/utils/config.py
code served (permissive licence) · get_code("1a594c7b9b550e5a")
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 ArgumentParser. 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 sys
import argparse

this_file_path = "/tmp/nlp_project/"
project_NLP_path = this_file_path

victim_model_version = 'bert_base_uncased'
steal_model_version = 'bert_base_uncased'

# Create a minimal argument parser for testing
parser = argparse.ArgumentParser()

ARGS = {
    'parser': parser
}
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 ArgumentParser — a ArgumentParser output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ArgumentParser 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
roberta_config Ran
ran; our contract draft was wrong, not the code
C-W-D/MeaeQ/utils/config.py
code served (permissive licence) · get_code("97bc105d51edbfd3")
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 ArgumentParser. 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 argparse
import sys

torch_imported = False
try:
    import torch
    torch.manual_seed(0)
    torch_imported = True
except ImportError:
    pass

this_file_path = "/fake/path/"

project_NLP_path = this_file_path

def roberta_config(parser):
    parser.add_argument('--roberta_hidden_size', type=int, default=768, help='')
    parser.add_argument('--roberta_hidden_dropout_prob', type=float, default=0.1, help='')
    parser.add_argument('--roberta_output_hidden_states', type=bool, default=True, help='')
    parser.add_argument('--steal_roberta_path', type=str, default=project_NLP_path + 'pretrained_model/' + 'roberta_base', help='')
    parser.add_argument('--steal_roberta_vocab_path', type=str, default=project_NLP_path + 'pretrained_model/' + 'roberta_base' + '/', help='')
    parser.add_argument('--rpc_dropout_prob', type=float, default=0.1, help='')
    parser.add_argument('--victim_roberta_path', type=str, default=project_NLP_path + 'pretrained_model/' + 'roberta_base', help='')
    parser.add_argument('--victim_roberta_vocab_path', type=str, default=project_NLP_path + 'pretrained_model/' + 'roberta_base' + '/', help='')
    return parser

parser = argparse.ArgumentParser()

ARGS = {
    'parser': parser
}
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 ArgumentParser — a ArgumentParser output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ArgumentParser 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
sentence_length_calculate Ran
ran, honoured the contract we drafted
C-W-D/MeaeQ/utils/tools.py
code served (permissive licence) · get_code("d55f86277b07ffd7")
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 int, ndim 0, dtype family int, 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

ARGS = {
    "sentence_list": [
        "hello world",
        "this is a longer sentence",
        "short",
        "another test sentence here",
        "x",
        "the quick brown fox jumps over the lazy dog"
    ],
    "alpha": 0.95
}
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 int, ndim 0, dtype family int, 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
to Ran
ran, honoured the contract we drafted
C-W-D/MeaeQ/utils/tools.py
code served (permissive licence) · get_code("0bfbcd38797df955")
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. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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 'to' takes a single parameter 'x' and moves it to GPU if available
# Since we're CPU-only, we just need to pass something that has a .cuda() method
# A tensor is the most straightforward choice

x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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
to_var Ran
ran; our contract draft was wrong, not the code
C-W-D/MeaeQ/utils/tools.py
code served (permissive licence) · get_code("8ea2da1113eabed8")
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. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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
from torch.autograd import Variable

torch.manual_seed(0)

# The function takes a single parameter 'x' which should be a tensor
# It checks if CUDA is available (it won't be in this sandbox)
# and returns a Variable wrapping the tensor

x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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
parse_arguments Not yet run
unverified (harvested, no recorded run)
C-W-D/MeaeQ/utils/config.py
code served (permissive licence) · get_code("dfba938414a5b52e")
How it ran

no invocation recorded.

split_steal_data Not yet run
unverified (harvested, no recorded run)
C-W-D/MeaeQ/steal/model_steal/al_steal.py
code served (permissive licence) · get_code("12b751ebfad841d8")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 0 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
to_var 4 other papers
4 dated earlier
1704.04394 (2017-04)
DESIRE: Distant Future Prediction in Dynamic Scenes with Interacting Agents
1905.01354 (2019-05)
Controllable Artistic Text Style Transfer via Shape-Matching GAN
2001.02890 (2020-01)
Deep Plastic Surgery: Robust and Controllable Image Editing with Human-Drawn Sketches
and 1 more not named here

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

We study model extraction attacks in natural language processing (NLP) where attackers aim to steal victim models by repeatedly querying the open Application Programming Interfaces (APIs). Recent works focus on limited-query budget settings and adopt random sampling or active learning-based sampling strategies on publicly available, unannotated data sources. However, these methods often result in selected queries that lack task relevance and data diversity, leading to limited success in achieving satisfactory results with low query costs. In this paper, we propose MeaeQ (Model extraction attack with efficient Queries), a straightforward yet effective method to address these issues. Specifically, we initially utilize a zero-shot sequence inference classifier, combined with API service information, to filter task-relevant data from a public text corpus instead of a problem domain-specific dataset. Furthermore, we employ a clustering-based data reduction technique to obtain representative data as queries for the attack. Extensive experiments conducted on four benchmark datasets demonstrate that MeaeQ achieves higher functional similarity to the victim model than baselines while requiring fewer queries. Our code is available at https://github.com/C-W-D/MeaeQ.

For agents

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

get_harvested_code_for_paper("2310.14047")
get_code_for_paper("2310.14047")
have("2310.14047")

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

curl https://syntology.ai/api/ran/2310.14047.json

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

[![Syntology run record](https://syntology.ai/api/ran/2310.14047.svg)](https://syntology.ai/paper/2310.14047)

Connect an agent — have() is free.