Prompting Large Language Models with Human Error Markings for Self-Correcting Machine Translation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914823256670208 |
|---|---|
| author | Berger, Nathaniel Riezler, Stefan Exel, Miriam Huck, Matthias |
| author_facet | Berger, Nathaniel Riezler, Stefan Exel, Miriam Huck, Matthias |
| contents | While large language models (LLMs) pre-trained on massive amounts of unpaired language data have reached the state-of-the-art in machine translation (MT) of general domain texts, post-editing (PE) is still required to correct errors and to enhance term translation quality in specialized domains. In this paper we present a pilot study of enhancing translation memories (TM) produced by PE (source segments, machine translations, and reference translations, henceforth called PE-TM) for the needs of correct and consistent term translation in technical domains.
We investigate a light-weight two-step scenario where, at inference time, a human translator marks errors in the first translation step, and in a second step a few similar examples are extracted from the PE-TM to prompt an LLM. Our experiment shows that the additional effort of augmenting translations with human error markings guides the LLM to focus on a correction of the marked errors, yielding consistent improvements over automatic PE (APE) and MT from scratch. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_02267 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Prompting Large Language Models with Human Error Markings for Self-Correcting Machine Translation Berger, Nathaniel Riezler, Stefan Exel, Miriam Huck, Matthias Computation and Language While large language models (LLMs) pre-trained on massive amounts of unpaired language data have reached the state-of-the-art in machine translation (MT) of general domain texts, post-editing (PE) is still required to correct errors and to enhance term translation quality in specialized domains. In this paper we present a pilot study of enhancing translation memories (TM) produced by PE (source segments, machine translations, and reference translations, henceforth called PE-TM) for the needs of correct and consistent term translation in technical domains. We investigate a light-weight two-step scenario where, at inference time, a human translator marks errors in the first translation step, and in a second step a few similar examples are extracted from the PE-TM to prompt an LLM. Our experiment shows that the additional effort of augmenting translations with human error markings guides the LLM to focus on a correction of the marked errors, yielding consistent improvements over automatic PE (APE) and MT from scratch. |
| title | Prompting Large Language Models with Human Error Markings for Self-Correcting Machine Translation |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2406.02267 |