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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Modern Large language models (LLMs) can still generate responses that may notbe aligned with human expectations or values. While many weight-based alignmentmethods have been proposed, many of them still leave models vulnerable toattacks when used on their own. To help mitigate this issue, we introduceBergeron, a framework designed to improve the robustness of LLMs againstadversarial attacks. Bergeron employs a two-tiered architecture. Here, asecondary LLM serves as a simulated conscience that safeguards a primary LLM.We do this by monitoring for and correcting potentially harmful text withinboth the prompt inputs and the generated outputs of the primary LLM. Empiricalevaluation shows that Bergeron can improve the alignment and robustness ofseveral popular LLMs without costly fine-tuning. It aids both open-source andblack-box LLMs by complementing and reinforcing their existing alignmenttraining.

Bergeron: Combating Adversarial Attacks through a Conscience-Based Alignment Framework