We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.CL

Change to browse by:

cs

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computation and Language

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

Abstract: 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.
Subjects: Computation and Language (cs.CL)
Journal reference: Association for the Advancement of Artificial Intelligence (AAAI) 2019
Cite as: arXiv:1811.00119 [cs.CL]
  (or arXiv:1811.00119v2 [cs.CL] for this version)

Submission history

From: Su Wang [view email]
[v1] Wed, 31 Oct 2018 21:04:04 GMT (274kb,D)
[v2] Wed, 14 Nov 2018 15:24:51 GMT (274kb,D)

Link back to: arXiv, form interface, contact.