Summary of Neural Machine Translation For Malayalam Paraphrase Generation, by Christeena Varghese et al.
Neural Machine Translation for Malayalam Paraphrase Generation
by Christeena Varghese, Sergey Koshelev, Ivan P. Yamshchikov
First submitted to arxiv on: 31 Jan 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This study investigates four methods for generating paraphrases in Malayalam by leveraging English paraphrasing resources and pre-trained Neural Machine Translation (NMT) models. The authors evaluate the resulting paraphrases using both automated metrics like BLEU, METEOR, and cosine similarity as well as human annotation. The findings indicate that automated evaluation measures may not be suitable for Malayalam, as they do not consistently align with human judgment. This discrepancy highlights the need for more nuanced paraphrase evaluation approaches, particularly for highly agglutinative languages. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This study tries to make machines better at rewriting text in Malayalam by using tools and models that work well for English. The researchers tested their methods and found out that the computer’s way of judging how good they are doesn’t always agree with what humans think is good. This shows us that we need new ways to check if machine-generated paraphrases are actually helpful, especially when working with languages like Malayalam. |
Keywords
» Artificial intelligence » Bleu » Cosine similarity » Translation
