# Towards Lossless Encoding of Sentences

## TL;DR

This paper introduces a near lossless encoding method for sentences that preserves original text features and allows reconstruction, improving text compression and retrieval for natural language processing tasks.

## Contribution

It proposes a novel near lossless encoding technique for sentences that enables reconstruction and feature richness, addressing a gap in text compression research.

## Key findings

- Effective encoding of sentences and sub-sequences demonstrated
- Good performance on sentiment analysis tasks
- Preserves original sequence information

## Abstract

A lot of work has been done in the field of image compression via machine learning, but not much attention has been given to the compression of natural language. Compressing text into lossless representations while making features easily retrievable is not a trivial task, yet has huge benefits. Most methods designed to produce feature rich sentence embeddings focus solely on performing well on downstream tasks and are unable to properly reconstruct the original sequence from the learned embedding. In this work, we propose a near lossless method for encoding long sequences of texts as well as all of their sub-sequences into feature rich representations. We test our method on sentiment analysis and show good performance across all sub-sentence and sentence embeddings.