English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The Machine Learning as a Service (MLaaS) market is rapidly expanding andbecoming more mature. For example, OpenAI's ChatGPT is an advanced largelanguage model (LLM) that generates responses for various queries withassociated fees. Although these models can deliver satisfactory performance,they are far from perfect. Researchers have long studied the vulnerabilitiesand limitations of LLMs, such as adversarial attacks and model toxicity.Inevitably, commercial ML models are also not exempt from such issues, whichcan be problematic as MLaaS continues to grow. In this paper, we discover a newattack strategy against LLM APIs, namely the prompt abstraction attack.Specifically, we propose Mondrian, a simple and straightforward method thatabstracts sentences, which can lower the cost of using LLM APIs. In thisapproach, the adversary first creates a pseudo API (with a lower establishedprice) to serve as the proxy of the target API (with a higher establishedprice). Next, the pseudo API leverages Mondrian to modify the user query,obtain the abstracted response from the target API, and forward it back to theend user. Our results show that Mondrian successfully reduces user queries'token length ranging from 13% to 23% across various tasks, including textclassification, generation, and question answering. Meanwhile, these abstractedqueries do not significantly affect the utility of task-specific and generallanguage models like ChatGPT. Mondrian also reduces instruction prompts' tokenlength by at least 11% without compromising output quality. As a result, theprompt abstraction attack enables the adversary to profit without bearing thecost of API development and deployment.

Mondrian: Prompt Abstraction Attack Against Large Language Models for Cheaper API Pricing