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

This study delves into the capabilities and limitations of Large LanguageModels (LLMs) in the challenging domain of conditional question-answering.Utilizing the Conditional Question Answering (CQA) dataset and focusing ongenerative models like T5 and UL2, we assess the performance of LLMs acrossdiverse question types. Our findings reveal that fine-tuned LLMs can surpassthe state-of-the-art (SOTA) performance in some cases, even without fullyencoding all input context, with an increase of 7-8 points in Exact Match (EM)and F1 scores for Yes/No questions. However, these models encounter challengesin extractive question answering, where they lag behind the SOTA by over 10points, and in mitigating the risk of injecting false information. A study withoracle-retrievers emphasizes the critical role of effective evidence retrieval,underscoring the necessity for advanced solutions in this area. Furthermore, wehighlight the significant influence of evaluation metrics on performanceassessments and advocate for a more comprehensive evaluation framework. Thecomplexity of the task, the observed performance discrepancies, and the needfor effective evidence retrieval underline the ongoing challenges in this fieldand underscore the need for future work focusing on refining training tasks andexploring prompt-based techniques to enhance LLM performance in conditionalquestion-answering tasks.

Towards leveraging LLMs for Conditional QA