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

Large Language Models (LLMs) have achieved unprecedented performance inNatural Language Generation (NLG) tasks. However, many existing studies haveshown that they could be misused to generate undesired content. In response,before releasing LLMs for public access, model developers usually align thoselanguage models through Supervised Fine-Tuning (SFT) or Reinforcement Learningwith Human Feedback (RLHF). Consequently, those aligned large language modelsrefuse to generate undesired content when facing potentially harmful/unethicalrequests. A natural question is "could alignment really prevent thoseopen-sourced large language models from being misused to generate undesiredcontent?''. In this work, we provide a negative answer to this question. Inparticular, we show those open-sourced, aligned large language models could beeasily misguided to generate undesired content without heavy computations orcareful prompt designs. Our key idea is to directly manipulate the generationprocess of open-sourced LLMs to misguide it to generate undesired contentincluding harmful or biased information and even private data. We evaluate ourmethod on 4 open-sourced LLMs accessible publicly and our finding highlightsthe need for more advanced mitigation strategies for open-sourced LLMs.

On the Safety of Open-Sourced Large Language Models: Does Alignment Really Prevent Them From Being Misused?