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

While large language models (LLMs) can answer many questions correctly, theycan also hallucinate and give wrong answers. Wikidata, with its over 12 billionfacts, can be used to ground LLMs to improve their factuality. This paperpresents WikiWebQuestions, a high-quality question answering benchmark forWikidata. Ported over from WebQuestions for Freebase, it consists of real-worlddata with SPARQL annotation. This paper presents a few-shotsequence-to-sequence semantic parser for Wikidata. We modify SPARQL to use theunique domain and property names instead of their IDs. We train the parser touse either the results from an entity linker or mentions in the query. Wefine-tune LLaMA by adding the few-shot training data to that used to fine-tuneAlpaca. Our experimental results demonstrate the effectiveness of thismethodology, establishing a strong baseline of 76% and 65% answer accuracy inthe dev and test sets of WikiWebQuestions, respectively. By pairing oursemantic parser with GPT-3, we combine verifiable results with qualified GPT-3guesses to provide useful answers to 96% of the questions in dev. We also showthat our method outperforms the state-of-the-art for the QALD-7 Wikidatadataset by 3.6% in F1 score.

Fine-tuned LLMs Know More, Hallucinate Less with Few-Shot Sequence-to-Sequence Semantic Parsing over Wikidata