Spinning Language Models: Risks of Propaganda-As-A-Service and Countermeasures
Open accessThe DOI matches its Crossref record; 33 of the 33 references it cites were found in a registry or an index.
Abstract
We investigate a new threat to neural sequence-to-sequence (seq2seq) models: training-time attacks that cause models to “spin” their outputs so as to support an adversary-chosen sentiment or point of view—but only when the input contains adversary-chosen trigger words. For example, a spinned1summarization model outputs positive summaries of any text that mentions the name of some individual or organization.Model spinning introduces a “meta-backdoor” into a model. Whereas conventional backdoors cause models to produce incorrect outputs on inputs with the trigger, outputs of spinned models preserve context and maintain standard accuracy metrics, yet also satisfy a meta-task chosen by the adversary.Model spinning enables propaganda-as-a-service, where propaganda is defined as biased speech. An adversary can create customized language models that produce desired spins for chosen triggers, then deploy these models to generate disinformation (a platform attack), or else inject them into ML training pipelines (a supply-chain attack), transferring malicious functionality to downstream models trained by victims.To demonstrate the feasibility of model spinning, we develop a new backdooring technique. It stacks an adversarial meta-task (e.g., sentiment analysis) onto a seq2seq model, backpropagates the desired meta-task output (e.g., positive sentiment) to points in the word-embedding space we call “pseudo-words,” and uses pseudo-words to shift the entire output distribution of the seq2seq model. We evaluate this attack on language generation, summarization, and translation models with different triggers and meta-tasks such as sentiment, toxicity, and entailment. Spinned models largely maintain their accuracy metrics (ROUGE and BLEU) while shifting their outputs to satisfy the adversary’s meta-task. We also show that, in the case of a supply-chain attack, the spin functionality transfers to downstream models.Finally, we propose a black-box, meta-task-independent defense that, given a list of candidate triggers, can detect models that selectively apply spin to inputs with any of these triggers.1We use “spinned” rather than “spun” to match how the word is used in public relations.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325
Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks
Reimers, N., & Gurevych, I. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1410 · Cited by 12,778 · cited 1×
“It injects candidate triggers into inputs from a test dataset, applies model θ∗ to the original and modified inputs, and uses Sentence-Transformers [64] to encode the resulting outputs into vectors.”
MethodExploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., & Liu, P. J. (2019) · arXiv · doi:10.48550/arxiv.1910.10683 · Cited by 10,929
Lewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., Stoyanov, V., & Zettlemoyer, L. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.703 · Cited by 8,821
Lewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., Stoyanov, V., & Zettlemoyer, L. (2019) · arXiv · doi:10.48550/arxiv.1910.13461 · Cited by 8,816
Get To The Point: Summarization with Pointer-Generator Networks
See, A., Liu, P. J., & Manning, C. D. (2017) · Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p17-1099 · Cited by 3,775
A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference
Williams, A., Nangia, N., & Bowman, S. (2018) · Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers) · doi:10.18653/v1/n18-1101 · Cited by 3,342
How to backdoor federated learning
Bagdasaryan, E., Veit, A., Hua, Y., Estrin, D., & Shmatikov, V. (2018) · arXiv · doi:10.48550/arxiv.1807.00459 · Cited by 1,495 · cited 2×
“Previous backdoor attacks on seq2seq models [3, 68, 82, 84] force the model to generate a predetermined sequence as part of its output when the input contains a trigger.”
BackgroundMethodSelf-citation
Cited by
Shen, X., Chen, Z., Backes, M., Shen, Y., & Zhang, Y. (2023) · arXiv · doi:10.48550/arxiv.2308.03825 · Cited by 861
Shen, X., Chen, Z., Backes, M., Shen, Y., & Zhang, Y. (2024) · Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3658644.3670388 · Cited by 441
MasterKey: Automated Jailbreak Across Multiple Large Language Model Chatbots
Deng, G., Liu, Y., Li, Y., Wang, K., Zhang, Y., Li, Z., Wang, H., Zhang, T., & Liu, Y. (2023) · arXiv · doi:10.48550/arxiv.2307.08715 · Cited by 418
“LLM Backdoor [10], [52], [26] and model hijacking [38], [41] attacks can also be broadly categorized under this type of assault.”
BackgroundFeng, S., Park, C. Y., Liu, Y., & Tsvetkov, Y. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2023.acl-long.656 · Cited by 149
MASTERKEY: Automated Jailbreaking of Large Language Model Chatbots
Deng, G., Liu, Y., Li, Y., Wang, K., Zhang, Y., Li, Z., Wang, H., Zhang, T., & Liu, Y. (2024) · Proceedings 2024 Network and Distributed System Security Symposium · doi:10.14722/ndss.2024.24188 · Cited by 137
Metadata
- Authors
- Bagdasaryan, E., & Shmatikov, V.
- Venue
- 2022 IEEE Symposium on Security and Privacy (SP)
- Publisher
- IEEE
- Published
- 2022-05
- Pages
- 769-786
- ISSN
- 1081-6011 (ISSN-L)
- OpenAlex
- W4288057743
- Cited by
- 36
- References
- 33
- FWCI
- 5.09
- Type
- conference paper
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Spinning Language Models: Risks of Propaganda-As-A-Service and Countermeasures. https://app.cochat.ai/papers/10.1109/sp46214.2022.9833572
- The paper
- Bagdasaryan, E., & Shmatikov, V. (2022). Spinning Language Models: Risks of Propaganda-As-A-Service and Countermeasures. 2022 IEEE Symposium on Security and Privacy (SP). https://doi.org/10.1109/sp46214.2022.9833572
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.