Papers › Defending Against Neural Fake News

Defending Against Neural Fake News

29 May 2019NeurIPS 2019 12arXiv:1905.12616archive 2025-07-28

Rowan Zellers, Ari Holtzman, Hannah Rashkin, Yonatan Bisk, Ali Farhadi, Franziska Roesner, Yejin Choi

Recent progress in natural language generation has raised dual-use concerns. While applications like summarization and translation are positive, the underlying technology also might enable adversaries to generate neural fake news: targeted propaganda that closely mimics the style of real news. Modern computer security relies on careful threat modeling: identifying potential threats and vulnerabilities from an adversary's point of view, and exploring potential mitigations to these threats. Likewise, developing robust defenses against neural fake news requires us first to carefully investigate and characterize the risks of these models. We thus present a model for controllable text generation called Grover. Given a headline like `Link Found Between Vaccines and Autism,' Grover can generate the rest of the article; humans find these generations to be more trustworthy than human-written disinformation. Developing robust verification techniques against generators like Grover is critical. We find that best current discriminators can classify neural fake news from real, human-written, news with 73% accuracy, assuming access to a moderate level of training data. Counterintuitively, the best defense against Grover turns out to be Grover itself, with 92% accuracy, demonstrating the importance of public release of strong generators. We investigate these results further, showing that exposure bias -- and sampling strategies that alleviate its effects -- both leave artifacts that similar discriminators can pick up on. We conclude by discussing ethical issues regarding the technology, and plan to release Grover publicly, helping pave the way for better detection of neural fake news.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 10 code samples harvested from 2 repositories linked to this paper: 6 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 7 samples from 1 repository, 4 ran (4 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 3 samples from 1 repository, 2 ran (1 with no instrument failure; 1 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

rowanz/grover officialmentioned on GitHubtfApache-2.0Syntology: 7 samples, 4 ran (0 instrument-failed) · 3 unverified report
HomuraNagato/NLP_finetune mentioned on GitHubpytorchSyntology: not harvested report
agermanidis/OpenGPT-2 mentioned on GitHubtfApache-2.0Syntology: 3 samples, 2 ran (1 instrument-failed) · 1 unverified report
srir-crypto/fake_news mentioned on GitHubtfApache-2.0Syntology: not harvested report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

10 samples harvested: 6 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · our draft was wrong
5ran
4unverified

Licence: 10 of the 10 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

rowanz/grover official repository 7 samples, 4 ran (0 instrument-failed) · 3 unverified
ind_where lm/validate.py ran Apache-2.0 recorded; this copy not marked cleared · pointer only · 1cfe416523aed73f · report fixture: signature array:ndarray/1/int, target:int/0/int, return_first_match:bool/0/bool, default_value:int/0/int · returned int · level recorded 2026-09-25
input_fn_builder in 13 papers lm/dataloader.py ran Apache-2.0 recorded; this copy not marked cleared · pointer only · 677632c738689822 · report fixture: signature input_files:list/1/str, seq_length:int/0/int, is_training:bool/0/bool, num_cpu_threads:int/0/int, evaluate_for_fixed_number_of_steps:bool/0/bool · returned function · level recorded 2026-09-25
mask_attention_for_ltr lm/modeling.py ran Apache-2.0 recorded; this copy not marked cleared · pointer only · c2dd26edab6b7b63 · report fixture: signature attention_scores:EagerTensor/4/float, attention_mask:EagerTensor/2/float · returned EagerTensor · level recorded 2026-09-25
reduce_rms lm/optimization_adafactor.py ran Apache-2.0 recorded; this copy not marked cleared · pointer only · 2d64405efb57086c · report fixture: signature x:EagerTensor/2/float · returned EagerTensor · level recorded 2026-09-25
classification_input_fn_builder lm/dataloader.py unverified Apache-2.0 recorded; this copy not marked cleared · pointer only · 3a13144afb00f6c4 · report
gelu in 308 papers
census: 43 implementations from 74 papers, 6 distinct outputs on one shared input
lm/utils.py unverified Apache-2.0 recorded; this copy not marked cleared · pointer only · 6558b9f62bb65244 · report
layer_norm in 32 papers lm/utils.py unverified Apache-2.0 recorded; this copy not marked cleared · pointer only · cc925f3f6a2fb18c · report
agermanidis/OpenGPT-2 community (archive-listed) 3 samples, 2 ran (1 instrument-failed) · 1 unverified
get_shape_list in 9 papers lm/utils.py ran · our draft was wrong fingerprinted Apache-2.0 recorded; this copy not marked cleared · pointer only · 35fdcad846dde14a · report fixture: signature tensor:EagerTensor/2/float, expected_rank:int/0/int, name:str/0/str · returned list · level recorded 2026-09-22
glob_dir runway_model.py ran fingerprinted Apache-2.0 recorded; this copy not marked cleared · pointer only · 2007d9bd9677694f · report fixture: signature path:str/0/str · returned list · level recorded 2026-09-24
create_initializer in 10 papers lm/modeling.py unverified Apache-2.0 recorded; this copy not marked cleared · pointer only · e3981aa2fadfb2eb · report

Tasks

Computer SecurityFake News DetectionText Generation

Datasets

Introduced by this paper, per the archive.

RealNews

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Fake News Detection Grover-Mega Grover-Mega Unpaired Accuracy 92.0% #2 of 5 Archive leaderboard report
Fake News Detection Grover-Mega Grover-Large Unpaired Accuracy 80.8% #3 of 5 Archive leaderboard report
Fake News Detection Grover-Mega BERT-Large Unpaired Accuracy 73.1% #4 of 5 Archive leaderboard report
Fake News Detection Grover-Mega GPT2 (355M) Unpaired Accuracy 70.1% #5 of 5 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections