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

Generative AI models face the challenge of hallucinations that can undermineusers' trust in such systems. We approach the problem of conversationalinformation seeking as a two-step process, where relevant passages in a corpusare identified first and then summarized into a final system response. This waywe can automatically assess if the answer to the user's question is present inthe corpus. Specifically, our proposed method employs a sentence-levelclassifier to detect if the answer is present, then aggregates thesepredictions on the passage level, and eventually across the top-ranked passagesto arrive at a final answerability estimate. For training and evaluation, wedevelop a dataset based on the TREC CAsT benchmark that includes answerabilitylabels on the sentence, passage, and ranking levels. We demonstrate that ourproposed method represents a strong baseline and outperforms a state-of-the-artLLM on the answerability prediction task.

Towards Reliable and Factual Response Generation: Detecting Unanswerable Questions in Information-Seeking Conversations