For this shared task, we have used several machine translation engines to produce translations (en ⇔ fr) by fine-tuning a dialog-oriented NMT engine and having NMT baseline translations post-edited with prompt engineering. Our objectives are to test the effectiveness of a fine-tuning strategy with help of a robust NMT model, to draw out a from-translation-to-post-editing pipeline, and to evaluate the strong and weak points of NMT systems.

This paper describes the MAKE-NMTVIZ Systems trained for the WMT 2023 Literary task. As a primary submission, we used Train, Valid1, test1 as part of the GuoFeng corpus (Wang et al., 2023) to fine-tune the mBART50 model with Chinese-English data. We followed very similar training parameters to (Lee et al. 2022) when fine-tuning mBART50. We trained for 3 epochs, using gelu as an activation function, with a learning rate of 0.05, dropout of 0.1 and a batch size of 16. We decoded using a beam search of size 5. For our contrastive1 submission, we implemented a fine-tuned concatenation transformer (Lupo et al., 2023). The training was developed in two steps: (i) a sentence-level transformer was implemented for 10 epochs trained using general, test1, and valid1 data (more details in contrastive2 system); (ii) second, we fine-tuned at document-level using 3-sentence concatenation for 4 epochs using train, test2, and valid2 data. During the fine-tuning, we used ReLU as an activation function, with an inverse square root learning rate, dropout of 0.1, and a batch size of 64. We decoded using a beam search of size. Four our contrastive2 and last submission, we implemented a sentence-level transformer model (Vaswani et al., 2017). The model was trained with general data for 10 epochs using general-purpose, test1, and valid 1 data. The training parameters were an inverse square root scheduled learning rate, a dropout of 0.1, and a batch size of 64. We decoded using a beam search of size 4. We then compared the three translation outputs from an interdisciplinary perspective, investigating some of the effects of sentence- vs document-based training. Computer scientists, translators and corpus linguists discussed the linguistic remaining issues for this discourse-level literary translation.

This paper examines some of the effects of prosodic boundaries on ASR outputs and Spoken Language Translations into English for two competing French structures (“c’est” dislocation vs. “c’est” parentheticals). One native speaker of French read 104 test sentences that were then submitted to two systems. We compared the outputs of two toolkits, SYSTRAN Pure Neural Server (SPNS9) (Crego et al., 2016) and Whisper. For SPNS9, we compared the translation of the text file used for the reading with the translation of the transcription generated through Vocapia ASR. We also tested the transcription engine for speech recognition uploading an MP3 file and used the same procedure for AI Whisper’s Web-scale Supervised Pretraining for Speech Recognition system (Radford et al., 2022). We reported WER for the transcription tasks and the BLEU scores for the different models. We evidenced the variability of the punctuation in the ASR outputs and discussed it in relation to the duration of the utterance. We discussed the effects of the prosodic boundaries. We described the status of the boundary in the speech-to-text systems, discussing the consequence for the neural machine translation of the rendering of the prosodic boundary by a comma, a full stop, or any other punctuation symbol. We used the reference transcript of the reading phase to compute the edit distance between the reference transcript and the ASR output. We also used textometric analyses with iTrameur (Fleury and Zimina, 2014) for insights into the errors that can be attributed to ASR or to Neural Machine translation.

In this paper, we discuss the improvements brought by the fine-tuning of mBART50 for the translation of a specific Farsi dataset of dislocations. Given our BLEU scores, our evaluation is mostly qualitative: we assess the improvements of our fine-tuning in the translations into French of our test dataset of Farsi. We describe the fine-tuning procedure and discuss the quality of the results in the translations from Farsi. We assess the sentences in the French translations that contain English tokens and for the English translations, we examine the ability of the fine- tuned system to translate Farsi dislocations into English without replicating the dislocated item as a double subject. We scrutinized the Farsi training data used to train for mBART50 (Tang et al., 2021). We fine-tuned mBART50 with samples from an in-house French-Farsi aligned translation of a short story. In spite of the scarcity of available resources, we found that fine- tuning with aligned French-Farsi data dramatically improved the grammatical well-formedness of the predictions for French, even if serious semantic issues remained. We replicated the experiment with the English translation of the same Farsi short story for a Farsi-English fine-tuning and found out that similar semantic inadequacies cropped up, and that some translations were worse than our mBART50 baseline. We showcased the fine-tuning of mBART50 with supplementary data and discussed the asymmetry of the situation, adding little data in the fine-tuning is sufficient to improve morpho-syntax for one language pair but seems to degrade translation to English.