# SEQ^3: Differentiable Sequence-to-Sequence-to-Sequence Autoencoder for Unsupervised Abstractive Sentence Compression

arXiv: https://arxiv.org/abs/1904.03651 · HTML: https://mlsanity.com/paper/1904.03651/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Apr 2019.

## Built on by (most cited first, 27 of 27)

- [Unlocking Compositional Generalization in Pre-trained Models Using Intermediate Representations](https://mlsanity.com/paper/2104.07478/) (2021)
- [ConvoSumm: Conversation Summarization Benchmark and Improved Abstractive Summarization with Argument Mining](https://mlsanity.com/paper/2106.00829/) (2021)
- [Civil Rephrases Of Toxic Texts With Self-Supervised Transformers](https://mlsanity.com/paper/2102.05456/) (2021)
- [Reason first, then respond: Modular Generation for Knowledge-infused Dialogue](https://mlsanity.com/paper/2111.05204/) (2021)
- [Impossible Distillation: from Low-Quality Model to High-Quality Dataset & Model for Summarization and Paraphrasing](https://mlsanity.com/paper/2305.16635/) (2023)
- [Neural Data-to-Text Generation with LM-based Text Augmentation](https://mlsanity.com/paper/2102.03556/) (2021)
- [Efficient Unsupervised Sentence Compression by Fine-tuning Transformers with Reinforcement Learning](https://mlsanity.com/paper/2205.08221/) (2022)
- [CiteSum: Citation Text-guided Scientific Extreme Summarization and Domain Adaptation with Limited Supervision](https://mlsanity.com/paper/2205.06207/) (2022)
- [Unsupervised Abstractive Opinion Summarization by Generating Sentences with Tree-Structured Topic Guidance](https://mlsanity.com/paper/2106.08007/) (2021)
- [Referee: Reference-Free Sentence Summarization with Sharper Controllability through Symbolic Knowledge Distillation](https://mlsanity.com/paper/2210.13800/) (2022)
- [A Character-Level Length-Control Algorithm for Non-Autoregressive Sentence Summarization](https://mlsanity.com/paper/2205.14522/) (2022)
- [Learning Non-Autoregressive Models from Search for Unsupervised Sentence Summarization](https://mlsanity.com/paper/2205.14521/) (2022)
- [Automatic Related Work Generation: A Meta Study](https://mlsanity.com/paper/2201.01880/) (2022)
- [Toward Unified Controllable Text Generation via Regular Expression Instruction](https://mlsanity.com/paper/2309.10447/) (2023)
- [Information-Theoretic Distillation for Reference-less Summarization](https://mlsanity.com/paper/2403.13780/) (2024)
- [Generating Multiple-Length Summaries via Reinforcement Learning for Unsupervised Sentence Summarization](https://mlsanity.com/paper/2212.10843/) (2022)
- [EASE: Extractive-Abstractive Summarization with Explanations](https://mlsanity.com/paper/2105.06982/) (2021)
- [Symbolic Autoencoding for Self-Supervised Sequence Learning](https://mlsanity.com/paper/2402.10575/) (2024)
- [Warmup Generations: A Task-Agnostic Approach for Guiding Sequence-to-Sequence Learning with Unsupervised Initial State Generation](https://mlsanity.com/paper/2502.12304/) (2025)
- [RTSUM: Relation Triple-based Interpretable Summarization with Multi-level Salience Visualization](https://mlsanity.com/paper/2310.13895/) (2023)
- [Paired by the Teacher: Turning Unpaired Data into High-Fidelity Pairs for Low-Resource Text Generation](https://mlsanity.com/paper/2509.25144/) (2025)
- [Text-Preserving Lossy Text Compression: A Study of Strategic Deletion and LLM Reconstruction](https://mlsanity.com/paper/2605.29000/) (2026)
- [Unsupervised Summarization Re-ranking](https://mlsanity.com/paper/2212.09593/) (2022)
- [Recent Trends in Unsupervised Summarization](https://mlsanity.com/paper/2305.11231/) (2023)
- [Unsupervised Candidate Answer Extraction through Differentiable Masker-Reconstructor Model](https://mlsanity.com/paper/2310.13106/) (2023)
- [From Lengthy to Lucid: A Systematic Literature Review on NLP Techniques for Taming Long Sentences](https://mlsanity.com/paper/2312.05172/) (2023)
- [MeloBottleneck: Self-Supervised Melody Skeleton Extraction with a Latent Subsequence Bottleneck](https://mlsanity.com/paper/2607.10233/) (2026)

## Abstract (arXiv)

Neural sequence-to-sequence models are currently the dominant approach in several natural language processing tasks, but require large parallel corpora. We present a sequence-to-sequence-to-sequence autoencoder (SEQ^3), consisting of two chained encoder-decoder pairs, with words used as a sequence of discrete latent variables. We apply the proposed model to unsupervised abstractive sentence compression, where the first and last sequences are the input and reconstructed sentences, respectively, while the middle sequence is the compressed sentence. Constraining the length of the latent word sequences forces the model to distill important information from the input. A pretrained language model, acting as a prior over the latent sequences, encourages the compressed sentences to be human-readable. Continuous relaxations enable us to sample from categorical distributions, allowing gradient-based optimization, unlike alternatives that rely on reinforcement learning. The proposed model does not require parallel text-summary pairs, achieving promising results in unsupervised sentence compression on benchmark datasets.