Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
184 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Word Alignment as Preference for Machine Translation (2405.09223v2)

Published 15 May 2024 in cs.CL and cs.AI

Abstract: The problem of hallucination and omission, a long-standing problem in machine translation (MT), is more pronounced when a LLM is used in MT because an LLM itself is susceptible to these phenomena. In this work, we mitigate the problem in an LLM-based MT model by guiding it to better word alignment. We first study the correlation between word alignment and the phenomena of hallucination and omission in MT. Then we propose to utilize word alignment as preference to optimize the LLM-based MT model. The preference data are constructed by selecting chosen and rejected translations from multiple MT tools. Subsequently, direct preference optimization is used to optimize the LLM-based model towards the preference signal. Given the absence of evaluators specifically designed for hallucination and omission in MT, we further propose selecting hard instances and utilizing GPT-4 to directly evaluate the performance of the models in mitigating these issues. We verify the rationality of these designed evaluation methods by experiments, followed by extensive results demonstrating the effectiveness of word alignment-based preference optimization to mitigate hallucination and omission. On the other hand, although it shows promise in mitigating hallucination and omission, the overall performance of MT in different language directions remains mixed, with slight increases in BLEU and decreases in COMET.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (45)
  1. Gpt-4 technical report. arXiv preprint arXiv:2303.08774.
  2. Neural machine translation by jointly learning to align and translate. In 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings.
  3. A multitask, multilingual, multimodal evaluation of chatgpt on reasoning, hallucination, and interactivity. In Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pages 675–718.
  4. Findings of the 2020 conference on machine translation (WMT20). In Proceedings of the Fifth Conference on Machine Translation, pages 1–55, Online. Association for Computational Linguistics.
  5. Findings of the 2017 conference on machine translation (WMT17). In Proceedings of the Second Conference on Machine Translation, pages 169–214, Copenhagen, Denmark. Association for Computational Linguistics.
  6. Language models are few-shot learners. In Advances in Neural Information Processing Systems, volume 33, pages 1877–1901. Curran Associates, Inc.
  7. Improving pretrained cross-lingual language models via self-labeled word alignment. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 3418–3430, Online. Association for Computational Linguistics.
  8. SpanAlign: Sentence alignment method based on cross-language span prediction and ILP. In Proceedings of the 28th International Conference on Computational Linguistics, pages 4750–4761, Barcelona, Spain (Online). International Committee on Computational Linguistics.
  9. Scaling instruction-finetuned language models. Journal of Machine Learning Research, 25(70):1–53.
  10. No language left behind: Scaling human-centered machine translation. arXiv preprint arXiv:2207.04672.
  11. Detecting and mitigating hallucinations in machine translation: Model internal workings alone do well, sentence similarity Even better. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 36–50, Toronto, Canada. Association for Computational Linguistics.
  12. HalOmi: A manually annotated benchmark for multilingual hallucination and omission detection in machine translation. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 638–653, Singapore. Association for Computational Linguistics.
  13. Chain-of-verification reduces hallucination in large language models. ArXiv, abs/2309.11495.
  14. Zi-Yi Dou and Graham Neubig. 2021. Word alignment by fine-tuning embeddings on parallel corpora. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, pages 2112–2128.
  15. A simple, fast, and effective reparameterization of ibm model 2. In Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 644–648.
  16. Language-agnostic BERT sentence embedding. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 878–891, Dublin, Ireland. Association for Computational Linguistics.
  17. SimCSE: Simple contrastive learning of sentence embeddings. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 6894–6910, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
  18. How good are gpt models at machine translation? a comprehensive evaluation. arXiv preprint arXiv:2302.09210.
  19. Lora: Low-rank adaptation of large language models. In International Conference on Learning Representations.
  20. SimAlign: High quality word alignments without parallel training data using static and contextualized embeddings. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 1627–1643, Online. Association for Computational Linguistics.
  21. Large language models are zero-shot reasoners. Advances in neural information processing systems, 35:22199–22213.
  22. Dual-alignment pre-training for cross-lingual sentence embedding. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3466–3478, Toronto, Canada. Association for Computational Linguistics.
  23. Enhancing cross-lingual sentence embedding for low-resource languages with word alignment. arXiv preprint arXiv:2404.02490.
  24. Detectgpt: Zero-shot machine-generated text detection using probability curvature. In International Conference on Machine Learning.
  25. A supervised word alignment method based on cross-language span prediction using multilingual bert. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 555–565.
  26. Franz Josef Och and Hermann Ney. 2003. A systematic comparison of various statistical alignment models. Computational linguistics, 29(1):19–51.
  27. Training language models to follow instructions with human feedback. In Advances in Neural Information Processing Systems, volume 35, pages 27730–27744. Curran Associates, Inc.
  28. Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pages 311–318, Philadelphia, Pennsylvania, USA. Association for Computational Linguistics.
  29. Language models are unsupervised multitask learners. OpenAI blog, 1(8):9.
  30. Direct preference optimization: Your language model is secretly a reward model. Advances in Neural Information Processing Systems, 36.
  31. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347.
  32. Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288.
  33. Modeling coverage for neural machine translation. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 76–85.
  34. Zephyr: Direct distillation of lm alignment. ArXiv, abs/2310.16944.
  35. Jannis Vamvas and Rico Sennrich. 2022. As little as possible, as much as necessary: Detecting over- and undertranslations with contrastive conditioning. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 490–500, Dublin, Ireland. Association for Computational Linguistics.
  36. Finetuned language models are zero-shot learners. In International Conference on Learning Representations.
  37. Zero-shot information extraction via chatting with chatgpt. ArXiv, abs/2302.10205.
  38. WSPAlign: Word alignment pre-training via large-scale weakly supervised span prediction. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 11084–11099, Toronto, Canada. Association for Computational Linguistics.
  39. PCL: Peer-contrastive learning with diverse augmentations for unsupervised sentence embeddings. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 12052–12066, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
  40. Taking notes on the fly helps language pre-training. In International Conference on Learning Representations.
  41. A paradigm shift in machine translation: Boosting translation performance of large language models. In The Twelfth International Conference on Learning Representations.
  42. Contrastive preference optimization: Pushing the boundaries of llm performance in machine translation. arXiv preprint arXiv:2401.08417.
  43. Siren’s song in the ai ocean: A survey on hallucination in large language models. ArXiv, abs/2309.01219.
  44. Veco 2.0: Cross-lingual language model pre-training with multi-granularity contrastive learning. arXiv preprint arXiv:2304.08205.
  45. Leveraging multi-lingual positive instances in contrastive learning to improve sentence embedding. In Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pages 976–991, St. Julian’s, Malta. Association for Computational Linguistics.
Citations (1)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets