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

Conversational question-answering (CQA) systems aim to create interactivesearch systems that effectively retrieve information by interacting with users.To replicate human-to-human conversations, existing work uses human annotatorsto play the roles of the questioner (student) and the answerer (teacher).Despite its effectiveness, challenges exist as human annotation istime-consuming, inconsistent, and not scalable. To address this issue andinvestigate the applicability of large language models (LLMs) in CQAsimulation, we propose a simulation framework that employs zero-shot learnerLLMs for simulating teacher-student interactions. Our framework involves twoLLMs interacting on a specific topic, with the first LLM acting as a student,generating questions to explore a given search topic. The second LLM plays therole of a teacher by answering questions and is equipped with additionalinformation, including a text on the given topic. We implement both the studentand teacher by zero-shot prompting the GPT-4 model. To assess the effectivenessof LLMs in simulating CQA interactions and understand the disparities betweenLLM- and human-generated conversations, we evaluate the simulated data fromvarious perspectives. We begin by evaluating the teacher's performance throughboth automatic and human assessment. Next, we evaluate the performance of thestudent, analyzing and comparing the disparities between questions generated bythe LLM and those generated by humans. Furthermore, we conduct extensiveanalyses to thoroughly examine the LLM performance by benchmarkingstate-of-the-art reading comprehension models on both datasets. Our resultsreveal that the teacher LLM generates lengthier answers that tend to be moreaccurate and complete. The student LLM generates more diverse questions,covering more aspects of a given topic.

Let the LLMs Talk: Simulating Human-to-Human Conversational QA via Zero-Shot LLM-to-LLM Interactions