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

Recent advances in prompt engineering enable large language models (LLMs) tosolve multi-hop logical reasoning problems with impressive accuracy. However,there is little existing work investigating the robustness of LLMs withfew-shot prompting techniques. Therefore, we introduce a systematic approach totest the robustness of LLMs in multi-hop reasoning tasks via domain-agnosticperturbations. We include perturbations at multiple levels of abstractions(e.g. lexical perturbations such as typos, and semantic perturbations such asthe inclusion of intermediate reasoning steps in the questions) to conductbehavioral analysis on the LLMs. Throughout our experiments, we find thatmodels are more sensitive to certain perturbations such as replacing words withtheir synonyms. We also demonstrate that increasing the proportion of perturbedexemplars in the prompts improves the robustness of few-shot prompting methods.

Noisy Exemplars Make Large Language Models More Robust: A Domain-Agnostic Behavioral Analysis