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 paper contains what the Georgetown InfoSense group has done in regard tosolving the challenges presented by TREC iKAT 2023. Our submitted runsoutperform the median runs by a significant margin, exhibiting superiorperformance in nDCG across various cut numbers and in overall success rate. Ourapproach uses a Generate-Retrieve-Generate method, which we've found to greatlyoutpace Retrieve-Then-Generate approaches for the purposes of iKAT. Oursolution involves the use of Large Language Models (LLMs) for initial answers,answer grounding by BM25, passage quality filtering by logistic regression, andanswer generation by LLMs again. We leverage several purpose-built LanguageModels, including BERT, Chat-based, and text-to-transfer-based models, for textunderstanding, classification, generation, and summarization. The officialresults of the TREC evaluation contradict our initial self-evaluation, whichmay suggest that a decrease in the reliance on our retrieval and classificationmethods is better. Nonetheless, our findings suggest that the sequence ofinvolving these different components matters, where we see an essentiality ofusing LLMs before using search engines.

Sequencing Matters: A Generate-Retrieve-Generate Model for Building Conversational Agents