Skip to content
AI News HubLIVE
Original source2 min read

RL-ATTACK: A Generalized Black-Box Adversarial Attack on Language Models via Reinforcement Learning

Summary

Researchers propose RL-ATTACK, a trainable reinforcement learning framework for black-box adversarial attacks that learns from the target model to generate tailored adversarial transformations. It drives post-attack accuracy to 2% within 20 queries on natural language inference and outperforms an untrained attacker and a strong optimization baseline in generative Q&A. The paper is accepted at IEEE EdgeCom 2026.

SourcearXiv Computational LinguisticsAuthor: Taikun Chen, Yue Zhang, Wei Qian, Hui Fang
RL-ATTACK: A Generalized Black-Box Adversarial Attack on Language Models via Reinforcement Learning
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[Submitted on 5 Nov 2026]

Title:RL-ATTACK: A Generalized Black-Box Adversarial Attack on Language Models via Reinforcement Learning

View a PDF of the paper titled RL-ATTACK: A Generalized Black-Box Adversarial Attack on Language Models via Reinforcement Learning, by Taikun Chen and 3 other authors

View PDF HTML (experimental)

Abstract:Large language models (LLMs) exhibit vulnerabilities to adversarial perturbations that can induce unintended responses. These vulnerabilities can be identified via intentional attacks and subsequently used to improve model safety. In black-box settings where gradients are inaccessible, existing methods tend to guess adversarial transformations using external knowledge, ignoring the information provided by the target model. To mitigate this, we propose a trainable black-box framework based on reinforcement learning, which learns from the target model to generate tailored adversarial transformations. We demonstrate that our approach achieves lower post-attack accuracy, greater query efficiency and better semantic preservation on both natural language inference and generative Q&A tasks. On natural language inference, it drives post-attack accuracy to 2% within 20 queries. Against generative language models, our trained attacker consistently achieves lower post-attack accuracy than the untrained attacker and wins in 4 out of 6 settings when compared to a strong optimization-based baseline. We further experiment with multiple reward designs and training settings, and we find that no design or setting is consistently dominant. Our code is available at this https URL.

Comments: Accepted as a regular paper at IEEE EdgeCom 2026

Subjects:

Computation and Language (cs.CL); Cryptography and Security (cs.CR)

Cite as: arXiv:2611.08711 [cs.CL]

(or arXiv:2611.08711v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2611.08711

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Taikun Chen [view email] [v1] Thu, 5 Nov 2026 21:06:21 UTC (960 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled RL-ATTACK: A Generalized Black-Box Adversarial Attack on Language Models via Reinforcement Learning, by Taikun Chen and 3 other authors

View PDF

HTML (experimental)

TeX Source

view license

Additional Features

Audio Summary

Current browse context:

cs.CL

new | recent | 2026-11

Change to browse by:

cs cs.CR

References & Citations

SciX

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

EngineersAdvanced

Key points

  • RL-ATTACK uses reinforcement learning to learn adversarial transformations from the target model instead of relying on external knowledge.
  • It reduces post-attack accuracy to 2% within 20 queries on natural language inference.
  • The trained attacker wins in 4 out of 6 settings against a strong optimization-based baseline on generative Q&A.
  • No single reward design or training setting is consistently dominant across tasks.

Highlights and analysis are generated automatically and may contain errors. Check the original source.