SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2202.09381 · AAAI · 2022

Synthetic Disinformation Attacks on Automated Fact Verification Systems

Christopher Manning, Antoine Bosselut, Yibing Du

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 1 has no contract verdict; 2 failures of our instrument; 2 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, 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/2202.09381.json · badge: /api/ran/2202.09381.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
yibing-du/adversarial-factcheck canonical (classifier label)
classifier basis: plan_row
5 ran · 2 no recorded run
FunctionStatusWhere it lives
to_list Ran
ran; our contract draft was wrong, not the code
yibing-du/adversarial-factcheck/fever/corefbert/Pretrain/run_pretrain.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from AI-secure/InfoBERT, is pointer only (licence: NONE) · get_code("9df40357afea56cc")
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 AI-secure/InfoBERT, 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.

kernal_mus Ran
ran, honoured the contract we drafted
yibing-du/adversarial-factcheck/fever/corefbert/FEVER/models.py
code served (permissive licence) · get_code("347bcdf9cb798d84")
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 torch
import numpy as np

ARGS = {
    "n_kernels": 5
}
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 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
kernel_sigmas Ran
ran, honoured the contract we drafted
yibing-du/adversarial-factcheck/fever/corefbert/FEVER/models.py
code served (permissive licence) · get_code("0bafc889ce780931")
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 torch
import numpy as np

torch.manual_seed(0)

ARGS = {
    "n_kernels": 5
}
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 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
tok2int_sent Ran
ran on a synthesized input
yibing-du/adversarial-factcheck/fever/corefbert/FEVER/data_loader.py
code served (permissive licence) · get_code("e6586b6790b3b3fc")
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, ndim 2, dtype family int; 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.

class MockTokenizer:
    def __init__(self):
        self.vocab = {
            '<s>': 0,
            '</s>': 1,
            'hello': 2,
            'world': 3,
            'this': 4,
            'is': 5,
            'a': 6,
            'test': 7,
            'title': 8,
            'sentence': 9,
        }
    
    def tokenize(self, text):
        if not text:
            return []
        words = text.lower().split()
        return words[:5]
    
    def convert_tokens_to_ids(self, tokens):
        ids = []
        for token in tokens:
            ids.append(self.vocab.get(token, 10))
        return ids


tokenizer = MockTokenizer()

sentence = (
    "hello world this is a test",
    "title",
    "this is another sentence"
)

max_seq_length = 32

ARGS = {
    "sentence": sentence,
    "tokenizer": tokenizer,
    "max_seq_length": max_seq_length,
}
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 tuple, ndim 2, dtype family int — a tuple 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
url_to_filename Ran
ran; our contract draft was wrong, not the code
yibing-du/adversarial-factcheck/fever/corefbert/FEVER/file_utils.py
code served (permissive licence) · get_code("07a2e3e996a6dc8e")
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 str. 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.

from hashlib import sha256

ARGS = {
    'url': 'https://example.com/model.bin',
    'etag': 'abc123def456'
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
load_and_cache_examples Not yet run
unverified (harvested, no recorded run)
yibing-du/adversarial-factcheck/fever/corefbert/Pretrain/run_pretrain.py
code served (permissive licence) · get_code("dc8e63d7478acf25")
How it ran

no invocation recorded.

tok2int_list Not yet run
unverified (harvested, no recorded run)
yibing-du/adversarial-factcheck/fever/corefbert/FEVER/data_loader.py
code served (permissive licence) · get_code("837f55825f82c177")
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, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 0 the same month, 2 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
to_list 41 other papers
28 dated earlier, 13 later
1902.01718 (2019-02)
End-to-End Open-Domain Question Answering with BERTserini
1910.06188 (2019-10)
Q8BERT: Quantized 8Bit BERT
1911.02984 (2019-11)
The TechQA Dataset
and 38 more not named here
url_to_filename 12 other papers
9 dated earlier, 3 later
1909.11942 (2019-09)
ALBERT: A Lite BERT for Self-supervised Learning of Language Representations
2004.09297 (2020-04)
MPNet: Masked and Permuted Pre-training for Language Understanding
2005.00796 (2020-05)
A Simple Language Model for Task-Oriented Dialogue
and 9 more not named here
kernal_mus 2 other papers
2 dated earlier
1910.09796 (2019-10)
Fine-grained Fact Verification with Kernel Graph Attention Network
2110.15064 (2021-10)
Towards Fine-Grained Reasoning for Fake News Detection
kernel_sigmas 2 other papers
2 dated earlier
1910.09796 (2019-10)
Fine-grained Fact Verification with Kernel Graph Attention Network
2110.15064 (2021-10)
Towards Fine-Grained Reasoning for Fake News Detection

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

Automated fact-checking is a needed technology to curtail the spread of online misinformation. One current framework for such solutions proposes to verify claims by retrieving supporting or refuting evidence from related textual sources. However, the realistic use cases for fact-checkers will require verifying claims against evidence sources that could be affected by the same misinformation. Furthermore, the development of modern NLP tools that can produce coherent, fabricated content would allow malicious actors to systematically generate adversarial disinformation for fact-checkers. In this work, we explore the sensitivity of automated factcheckers to synthetic adversarial evidence in two simulated settings: ADVERSARIAL ADDITION, where we fabricate documents and add them to the evidence repository available to the fact-checking system, and ADVERSARIAL MODIFICATION, where existing evidence source documents in the repository are automatically altered. Our study across multiple models on three benchmarks demonstrates that these systems suffer significant performance drops against these attacks. Finally, we discuss the growing threat of modern NLG systems as generators of disinformation in the context of the challenges they pose to automated fact-checkers.

For agents

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

get_harvested_code_for_paper("2202.09381")
get_code_for_paper("2202.09381")
have("2202.09381")

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

curl https://syntology.ai/api/ran/2202.09381.json

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

[![Syntology run record](https://syntology.ai/api/ran/2202.09381.svg)](https://syntology.ai/paper/2202.09381)

Connect an agent — have() is free.