CX DB8: A queryable extractive summarizer and semantic search engine

TL;DR

CX_DB8 is a novel unsupervised extractive summarizer and semantic search engine tailored for competitive debate, enabling biasable, rapid summarization and evidence retrieval from large texts, with improvements aligned to underlying embedding models.

Contribution

The paper introduces CX_DB8, a new framework combining biasable extractive summarization and semantic search using pre-trained text embeddings, enhancing evidence production tools for debate.

Findings

CX_DB8 effectively summarizes texts with bias towards target meanings.

It functions as a semantic search engine for evidence retrieval.

The framework improves as underlying embedding models evolve.

Abstract

Competitive Debate's increasingly technical nature has left competitors looking for tools to accelerate evidence production. We find that the unique type of extractive summarization performed by competitive debaters - summarization with a bias towards a particular target meaning - can be performed using the latest innovations in unsupervised pre-trained text vectorization models. We introduce CX_DB8, a queryable word-level extractive summarizer and evidence creation framework, which allows for rapid, biasable summarization of arbitarily sized texts. CX_DB8s usage of the embedding framework Flair means that as the underlying models improve, CX_DB8 will also improve. We observe that CX_DB8 also functions as a semantic search engine, and has application as a supplement to traditional "find" functionality in programs and webpages. CX_DB8 is currently used by competitive debaters and is made available to the public at https://github.com/Hellisotherpeople/CX_DB8