# A task in a suit and a tie: paraphrase generation with semantic augmentation

arXiv: https://arxiv.org/abs/1811.00119 · HTML: https://mlsanity.com/paper/1811.00119/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Oct 2018.

## Built on by (most cited first, 15 of 15)

- [ConRPG: Paraphrase Generation using Contexts as Regularizer](https://mlsanity.com/paper/2109.00363/) (2021)
- [TOD-DA: Towards Boosting the Robustness of Task-oriented Dialogue Modeling on Spoken Conversations](https://mlsanity.com/paper/2112.12441/) (2021)
- [Measuring and Modifying the Readability of English Texts with GPT-4](https://mlsanity.com/paper/2410.14028/) (2024)
- [Retrieve & Memorize: Dialog Policy Learning with Multi-Action Memory](https://mlsanity.com/paper/2106.02317/) (2021)
- [Pushing Paraphrase Away from Original Sentence: A Multi-Round Paraphrase Generation Approach](https://mlsanity.com/paper/2109.01862/) (2021)
- [Simulated annealing for optimization of graphs and sequences](https://mlsanity.com/paper/2110.01384/) (2021)
- [Weakly Supervised Reasoning by Neuro-Symbolic Approaches](https://mlsanity.com/paper/2309.13072/) (2023)
- [Improving the Diversity of Unsupervised Paraphrasing with Embedding Outputs](https://mlsanity.com/paper/2110.13231/) (2021)
- [Coherence and Diversity through Noise: Self-Supervised Paraphrase Generation via Structure-Aware Denoising](https://mlsanity.com/paper/2302.02780/) (2023)
- [Adverbs, Surprisingly](https://mlsanity.com/paper/2305.19650/) (2023)
- [Task-Oriented Paraphrase Analytics](https://mlsanity.com/paper/2403.17564/) (2024)
- [Paraphrase Generation as Unsupervised Machine Translation](https://mlsanity.com/paper/2109.02950/) (2021)
- [Enhancing Personality Recognition in Dialogue by Data Augmentation and Heterogeneous Conversational Graph Networks](https://mlsanity.com/paper/2401.05871/) (2024)
- [Action Controlled Paraphrasing](https://mlsanity.com/paper/2405.11277/) (2024)
- [SMCLM: Semantically Meaningful Causal Language Modeling for Autoregressive Paraphrase Generation](https://mlsanity.com/paper/2507.03415/) (2025)

## Abstract (arXiv)

Paraphrasing is rooted in semantics. We show the effectiveness of transformers (Vaswani et al. 2017) for paraphrase generation and further improvements by incorporating PropBank labels via a multi-encoder. Evaluating on MSCOCO and WikiAnswers, we find that transformers are fast and effective, and that semantic augmentation for both transformers and LSTMs leads to sizable 2-3 point gains in BLEU, METEOR and TER. More importantly, we find surprisingly large gains on human evaluations compared to previous models. Nevertheless, manual inspection of generated paraphrases reveals ample room for improvement: even our best model produces human-acceptable paraphrases for only 28% of captions from the CHIA dataset (Sharma et al. 2018), and it fails spectacularly on sentences from Wikipedia. Overall, these results point to the potential for incorporating semantics in the task while highlighting the need for stronger evaluation.