Pith. sign in

REVIEW 6 major objections 9 minor 25 references

Systematic Analysis of Large Language Models and Transformer-Based Machine Translation for English-Tamil and Tamil-English Across Diverse Datasets

T0 review · 6 major / 9 minor · reviewed 2026-07-31 · grok-4.5

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read For English–Tamil translation, parallel-data quality and domain alignment dominate model scores more than which off-the-shelf multilingual system you pick.

desk verdict Useful English–Tamil multi-dataset bake-off, but the PMIndia near-zeros look like a broken eval pipeline, not clean proof that noise dominates. read the letter →

arxiv 2607.24515 v2 pith:6TMUFVAM submitted 2026-07-27 cs.CL

Sriharshaa S , Sangeetha Sivanesan , Jaya Nirmala S This is my paper · ORCID
classification cs.CL
keywords English-Tamilmachinetranslationlow-resourceNMTNLLBmBARTattentionentropydatasetqualityfew-shotLLMchrF
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

English–Tamil and Tamil–English remain hard for neural machine translation because parallel data are scarce, often noisy, and domains shift, while Tamil’s morphology is rich. This paper runs two widely used multilingual transformers, NLLB and mBART, off the shelf on four parallel sets that range from clean (EnTamV2) to severely mismatched (PMIndia), scoring them with BLEU and chrF and reading their cross-attention. It finds that well-aligned data give stable, moderate gains and that very noisy data drive scores near zero, while lower attention entropy tracks better quality. A Tamil-adapted LLaMA used few-shot also produces fluent conversational output, though it is less literal than the supervised systems on harder sentences. A sympathetic reader cares because the result says where to spend effort first: clean and match the data before chasing larger models.

What carries the argument

Cross-attention heatmaps and sentence-level attention entropy H(y) = −∑ aᵢⱼ log(aᵢⱼ) on the decoder’s final layer, used as a quantitative proxy for how sharply source tokens align to target tokens and how that sharpness tracks BLEU/chrF across clean versus noisy corpora.

What would settle it

Fine-tune NLLB and mBART on equal amounts of cleaned EnTamV2-style versus PMIndia-style data under the same recipe, then check whether the huge metric gap shrinks or vanishes; if it vanishes, dataset quality is not the dominant driver under the paper’s setup.

Watch

Extended reading notes

Core claim

Across English–Tamil and Tamil–English, dataset noise level and alignment quality largely determine the BLEU and chrF of off-the-shelf NLLB and mBART: clean, well-aligned sets such as EnTamV2 yield consistent moderate scores, while highly noisy, domain-mismatched sets such as PMIndia yield near-zero scores, and lower cross-attention entropy co-occurs with better automatic quality. Few-shot TamilLaMA can still produce structurally coherent, fluent translations, especially in conversational settings.

Load-bearing premise

Score gaps can be blamed mainly on dataset noise and alignment when the models are only run with decoding tweaks and never fully fine-tuned or trained under matched conditions on each corpus.

Editorial extensions

If this is right

  • Cleaning and domain-matching English–Tamil parallel data should raise BLEU/chrF more reliably than swapping among off-the-shelf multilingual checkpoints alone.
  • Attention entropy can serve as a cheap diagnostic for whether a corpus is too noisy or misaligned before large training runs.
  • chrF should be reported alongside BLEU for Tamil, because inflated BLEU with unstable chrF flags misalignment (as on WikiMatrix).
  • Few-shot Tamil-capable LLMs are usable for fluent conversational translation but remain weaker than supervised NMT on literal, entity-heavy, or long sentences.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Corpus-quality gates and automatic noise filters may be higher-leverage investments for Indic MT than scaling another generic multilingual model.
  • The same entropy-versus-quality pattern could be tested as a cheap filter when mining new web parallel data for other agglutinative low-resource languages.
  • Hybrid pipelines that use supervised NMT for literal fidelity and few-shot LLMs for conversational polish are a natural next experiment the paper leaves open.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

6 major / 9 minor

Summary. The manuscript evaluates two off-the-shelf multilingual NMT models (NLLB, mBART) and a few-shot prompted Tamil-adapted LLaMA on English↔Tamil translation across four parallel corpora of differing quality (EnTamV2, NTREX, WikiMatrix, PMIndia), using BLEU and chrF, plus a cross-attention interpretability analysis with an attention-entropy measure. The central claim is that dataset quality and domain alignment dominate performance: well-aligned EnTamV2 yields consistent gains, while PMIndia allegedly yields near-zero scores; lower attention entropy is claimed to track translation quality; few-shot TamilLaMA is claimed to produce fluent conversational translations while lagging on literal accuracy. The directional finding (clean, well-aligned data beats noisy, mismatched data) is plausible and consistent with most of Table 2, but the paper's single strongest evidence row — PMIndia — exhibits chrF ≈ 0.5, which implies near-zero character overlap and is the signature of a preprocessing/alignment/encoding failure rather than corpus noise; the paper's own corrupted Tamil strings and its duplicated PB-SMT baseline number reinforce this concern. Several results (WikiMatrix mBART 28.21 BLEU vs NLLB 9.88 on identical data) are anomalous and unexplained because checkpoints, preprocessing, and decoding details are undisclosed.

Significance. Systematic, reproducible benchmarking for genuinely low-resource pairs like English–Tamil is valuable to the community, and the paper attempts something broader than a single-model report: per-dataset comparison, an attention-entropy diagnostic, and a few-shot LLM comparison. If the PMIndia anomaly is diagnosed and the checkpoints/preprocessing are disclosed, the finding that off-the-shelf multilingual NMT performance tracks corpus provenance and domain would be a useful, falsifiable empirical result. However, the current version's headline numbers do not yet support its conclusions: its strongest evidence row has the signature of a pipeline failure, and no error bars, significance tests, checkpoint details, or reproducible decoding configurations are provided. The contribution is presently closer to an exploratory study than a systematic analysis.

major comments (6)
  1. [Table 2, PMIndia row; §4.1.3; Conclusion] The PMIndia row is the load-bearing evidence for the central claim ('near-zero scores when using a noisy dataset like PMIndia', Conclusion), and it is far more consistent with an evaluation/preprocessing failure than with corpus noise. chrF of 0.44–1.14 implies essentially zero character n-gram overlap between hypothesis and reference; even a severely domain-mismatched model emitting grammatical text in the correct language and script virtually always exceeds chrF 15–25. Three internal signals point to an artifact: (i) the same table reports PB-SMT on PMIndia at BLEU 9.56 (Thapliyal et al., 2022), so the corpus is demonstrably usable; PMIndia is human-translated government text — domain-mismatched, yes, but not 'very high noise' in the mined-corpus sense; (ii) the Tamil strings quoted in §4.2.2 and §4.3.2 contain visibly corrupted/mixed-script glyphs (e.g., 'அரா ழராமப்௖ற பஶொக௬க்கான'), in
  2. [Table 2, WikiMatrix and PMIndia rows] The PB-SMT (Thapliyal et al., 2022) baseline is listed at exactly BLEU 9.56 for both the WikiMatrix row and the PMIndia row. This duplication is almost certainly a copy error — the two corpora are distinct — and one or both attributions are wrong. Separately, Table 2 mixes prior-work numbers (MIDAS, Transformer-NMT, PB-SMT) with the authors' own scores without stating whether tokenization, BLEU implementation, or test splits are comparable; Post (2018) is cited but no sacreBLEU signature is given. The baseline rows must be corrected and comparability conditions stated, since several discussion claims (§4.1.1, §4.1.3) lean on these comparisons.
  3. [Table 2, WikiMatrix row; §3.3] On WikiMatrix EN→TA, mBART is reported at BLEU 28.21 versus NLLB at 9.88 on identical data — a 3× gap between two off-the-shelf multilingual models — and TA→EN scores of 31–40 BLEU are far above published off-the-shelf numbers for this pair on noisy mined data. The paper never states which checkpoints were used (mBART-base vs mBART-50, and which fine-tuned direction), which language tags were forced, or the exact preprocessing. A gap this large strongly suggests a configuration difference (e.g., a fine-tuned mBART-50-many-to-one checkpoint versus a mismatched NLLB language tag), not a model-quality difference. Checkpoint IDs, decoding hyperparameters (beam size, length penalty values), and tokenization must be reported; otherwise neither the WikiMatrix anomaly nor the cross-model comparison is interpretable.
  4. [§3.3; §4.1.3; Table 3] The study runs NLLB and mBART strictly in inference mode ('without performing full fine-tuning but only decoding optimization', §3.3), yet attributes the cross-dataset score gaps primarily to dataset noise and alignment (§4.1.3, Table 3, Conclusion). With inference-only evaluation, score differences confound corpus noise with (a) the match between each corpus's domain and the models' pretraining/fine-tuning data (NLLB was trained partly on NLLB-Seed and WikiMatrix-adjacent mined data) and (b) unstated checkpoint differences. The noise-level labels in Table 3 are themselves qualitative and partly circular: WikiMatrix is labeled 'High noise / Low alignment' yet yields the highest BLEU/chrF scores in the whole study, while PMIndia — human-translated — is labeled 'Very High noise' on the basis of the anomalous near-zero row. At minimum the claims must be hedged to 'off-the-shelf performance
  5. [§4.3.3–4.3.4; Figure 6; Conclusion] §4.3.4 claims 'a consistent relation between the attention entropy and the translation performance' and the Conclusion elevates entropy to 'a quantitative way to measure the alignment of the dataset.' But (i) no entropy numbers are reported anywhere — Figure 6 is the only evidence and no correlation coefficient or per-dataset means are given; (ii) cross-attention entropy is a property of the model's decoding behavior, not of the dataset, so treating it as a dataset-alignment measure conflates model uncertainty with data noise; (iii) the literature (e.g., the attention-is-not-explanation line of work) cautions against reading alignment directly off attention weights, a caveat the paper does not acknowledge despite citing Ghader & Monz (2017). The section needs actual entropy values per dataset, a reported correlation with BLEU/chrF (with n and p), and reframed language; as written the int
  6. [§4.2; Table 4] The TamilLaMA experiments are underspecified to the point of irreproducibility, and contain an internal contradiction: §4.2.1 says the model 'has been specifically fine-tuned/adapted' to Tamil, while §4.2.2 says 'No fine-tuning was done.' The checkpoint/adapter used, number of few-shot exemplars, exemplar selection procedure, prompt template, and generation settings are all unstated. Table 4 mixes previously reported IN22 numbers with the authors' own 'IN22 (evaluated)' rows without stating whether the evaluation split, reference set, and metric implementation match — so the claim that TamilLaMA is competitive with GPT-3.5/LLaMA-2-13B is not on a controlled footing. Additionally, the few-shot results are evaluated qualitatively only (Limitations acknowledges this), yet the abstract claims in-context prompting is 'an excellent way' to perform translation. The setup must be fully specified
minor comments (9)
  1. [§4.3.2, Figures 4–5] Figure 4 is captioned 'Attention HeatMap of Tamil to English Translation' but the token mapping given underneath lists an English source (x₂ = The … x₇ = policies) and a Tamil target — i.e., English→Tamil. Figure 5 has the reverse inconsistency. Captions or mappings need swapping.
  2. [Throughout] Dataset names are inconsistently spelled throughout: 'EnTamV2'/'EntamV2'/'En TamV2', 'NTREX'/'NTrex'/'NTrex Benchmark', 'WikiMatrix'/'Wiki Matrix'/'WikiMatri x' (Table 2), 'TamilLaMA'/'Tamil LaMA'/'Tamil Lama'. Please standardize.
  3. [§4.1.2] §4.1.2 claims decoding optimization yielded 'increases of around 0.5–1.0 BLEU points and a chrF score between 1 and 2 points,' but no before/after numbers are given. Either include a small ablation table (baseline decoding vs tuned decoding) or remove the quantitative claim.
  4. [§3.4; §4.1] No statistical reliability information is provided: single decoding run per configuration, no confidence intervals or significance tests (e.g., bootstrap resampling for BLEU/chrF), and test-set sizes per dataset are never stated. At minimum, report test-set sizes and sacreBLEU confidence intervals.
  5. [Figures 2–6; §4.3.1] The heatmaps in Figures 2–6 are described qualitatively ('clear and nearly diagonal', 'fuzzy and irregular') but the figures are small and unlabeled in the preprint; axis labels (token identities), colorbars, and the averaging scheme over layers/heads should be specified. Note also that averaging or selecting the 'final layer' (§4.3.1) materially changes attention patterns; justify the choice.
  6. [§3.4] chrF is described as measuring 'semantic similarity at a character level' (§3.4) — it measures character n-gram overlap, not semantics. Also, chrF version (chrF vs chrF++) and the BLEU tokenization should be stated for reproducibility.
  7. [References] Several references are incomplete or missing venue/year details: Thillainathan et al. (2025), Raja & Vats (2025), Lakew et al. (2018), Hendy et al. (2023), Jiao et al. (2023) lack full bibliographic data; Gangadharaiah et al. (2023) lacks a venue. The NTREX dataset itself is used but never cited (Rikters et al. / the Microsoft NTREX-128 release).
  8. [Throughout] Typos and grammar: 'explain ability' (Abstract); 'This performs an attention-based analysis' (Abstract, missing subject); 'multi NMT models' (§1.1); 'the behavior of the translates' (§4.3); 'rep resents' (§4.3.3); 'Tamil capable TamilLaMA' (Abstract). A careful proofreading pass is warranted.
  9. [§4.2.2] The qualitative examples in §4.2.2 are used to claim TamilLaMA is 'more fluent and natural,' but with two cherry-picked sentences and no annotator protocol this cannot support even a qualitative conclusion. Either add a small structured human evaluation (raters, criteria, agreement) or soften the claim to an illustrative observation.

Circularity Check

0 steps flagged · score 0.0 of 10

Empirical MT benchmarking with no derivation chain; results are measured outputs, not quantities forced by construction.

full rationale

The paper is a systematic evaluation of off-the-shelf NLLB and mBART (plus qualitative few-shot TamilLaMA) on external English–Tamil corpora, reporting standard BLEU/chrF and attention diagnostics. There is no claimed first-principles derivation, fitted parameter renamed as a prediction, uniqueness theorem, or load-bearing self-citation chain. Dataset-quality conclusions rest on observed score gaps across named public sets (Tables 2–3); attention entropy (Eq. 1) is a post-hoc diagnostic correlated with those scores, not defined to equal them. Table 3’s qualitative “noise/alignment” labels track the metrics but do not mathematically force the reported numbers. No circular step meets the quote-and-reduction standard.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

Load-bearing commitments are standard MT evaluation assumptions plus author judgments about corpus noise and the decision to evaluate frozen multilingual checkpoints. No new physical entities; free choices are decoding and analysis hyperparameters that are only partly reported.

free parameters (3)
  • beam_size_and_length_normalization = unspecified; gains cited as ~0.5–1.0 BLEU, 1–2 chrF
    Decoding hyperparameters said to add ~0.5–1.0 BLEU and 1–2 chrF; exact values not given but claimed to drive reported gains (§4.1.2).
  • dataset_noise_alignment_labels = categorical labels per dataset
    Table 3 assigns Low/Medium/High/Very High noise and alignment grades that organize the central narrative; these are qualitative author labels, not independent measurements.
  • few_shot_exemplar_set_and_prompt
    In-context examples and prompt wording for TamilLaMA are chosen by hand and only illustrated, not fixed in a released template (§4.2.2).
assumptions (4)
  • domain assumption BLEU and chrF against given references are adequate automatic proxies for translation quality on morphologically rich Tamil, with chrF preferred when morphology matters.
    Stated in §3.4 and used as sole automatic evidence for NMT claims; no human eval on main grids.
  • domain assumption Cross-attention weights from the final decoder layer (and their entropy) meaningfully reflect source–target alignment quality for interpretability.
    §3.5 and §4.3; standard but contested assumption in NMT interpretability literature the paper cites.
  • domain assumption Public dataset releases (NTREX, EnTamV2, WikiMatrix, PMIndia) as used here are comparable test conditions once normalized and subword-tokenized.
    §3.1–3.2; ignores possible split leakage, filtering differences, and script normalization edge cases.
  • ad hoc to paper Inference-only multilingual checkpoints without task fine-tuning still license dataset-quality conclusions for this language pair.
    Explicit choice in §3.3; central to attributing failures to data rather than adaptation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Systematic Analysis of Large Language Models and Transformer-Based Machine Translation for English-Tamil and Tamil-English Across Diverse Datasets." pith.science (2026). https://pith.science/paper/6TMUFVAM

@misc{pith2026260724515,
  author       = {Pith},
  title        = {Pith review of: Systematic Analysis of Large Language Models and Transformer-Based Machine Translation for English-Tamil and Tamil-English Across Diverse Datasets},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6TMUFVAM}},
  note         = {Machine review of arXiv:2607.24515}
}
read the original abstract

The challenge of Machine Translation for low resource languages such as Tamil is primarily caused by the restricted amount of parallel data for these languages, as well as their substantial amount of domain variation and morphological complexity. This research presents the comprehensive evaluation of the performance of several multilingual translation models on English-Tamil and Tamil-English translations across multiple datasets: NTREX, EnTamV2, WikiMatrix and PMIndia. This study evaluates supervised NMT systems, NLLB and mBART, using both the BLEU and chrF metric, and examines how these systems perform on data of different quality levels and domains. This performs an attention-based analysis to increase model interpretability by visualising the alignments of tokens in an English source text and their Tamil translations and vice-versa to provide insight into how they make translations. This study also demonstrates that using in-context prompting can provide an excellent way to perform a few-shot translation of English to Tamil and Tamil-English using a Tamil capable TamilLaMA model, and compare this to supervised approaches qualitatively. These findings show that the quality of the datasets and their alignment with the domain will greatly affect the performance of the model, that attention-based mechanisms can aid in explain ability, and that few-shot large language models can still produce structurally coherent translations of Tamil.

Figures

Figures reproduced from arXiv: 2607.24515 by the authors.

Figure 1
Figure 1. System Architecture Diagram 4. Results and Discussion Experimental results are taken using transformer based neural machine translation systems and few shot prompting of a large language model for English to Tamil translation.The performance of the English to Tamil translation system has been evaluated using the following standard measures: Automatic Evaluation (BLEU and chrF), Qualitative Evaluation and Interpretab… view at source ↗
Figure 2
Figure 2. Attention HeatMap of EntamV2 dataset [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Heatmap illustrating attention patterns in the WikiMatrix corpus [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Attention HeatMap of Tamil to English Translation [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]
Figure 5
Figure 5. Figure 5: Attention HeatMap of English to Tamil Translation [PITH_FULL_IMAGE:figures/full_fig_p011_5.png]
Figure 6
Figure 6. Figure 6: Attention Entropy across Datasets 5. Conclusion In this study, systematic comparative evaluation of transformer-based multilingual models is provided – NLLB and mBART – in English to Tamil and Tamil to English translations using four different datasets with varying qua…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

25 extracted references

  1. [1]

    Himanshu Choudhary, Aditya Kumar Pathak, Rajiv Ratn Shah, and Ponnurangam Kumaraguru. 2018. Neural Machine Translation for English–Tamil. In Proceedings of the Third Conference on Machine Translation (WMT 2018), pages 770–775

    2018

  2. [2]

    McCrae, and Paul Buitelaar

    Rashmi Gangadharaiah, Bharathi Raja Chakravarthi, John P. McCrae, and Paul Buitelaar. 2023. EnTamV2: Enhanced Tamil–English Translation Dataset

    2023

  3. [3]

    Gomez, Łukasz Kaiser, and Illia Polosukhin

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention Is All You Need. In Proceedings of NeurIPS 2017

    2017

  4. [4]

    Vishal Thapliyal, Anoop Kunchukuttan, and Pushpak Bhattacharyya. 2022. PMIndia: A Parallel Corpus for Indian Languages

    2022

  5. [5]

    Rico Sennrich, Barry Haddow, and Alexandra Birch. 2016. Neural Machine Translation of Rare Words with Subword Units. In Proceedings of ACL 2016

    2016

  6. [6]

    Yinhan Liu, Jiatao Gu, Naman Goyal, Xian Li, Sergey Edunov, et al. 2020. Multilingual Denoising Pre- training for Neural Machine Translation. TACL

    2020

  7. [7]

    Naman Goyal, Cynthia Gao, Vishrav Chaudhary, Peng-Jen Chen, Guillaume Wenzek, et al. 2022. No Language Left Behind: Scaling Human-Centered Machine Translation. ACL 2022

    2022

  8. [8]

    Jörg Tiedemann and Santhosh Thottingal. 2020. OPUS-MT — Building Open Translation Services for the World. EAMT 2020

    2020

Show all 25 references
  1. [9]

    Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. BLEU: A Method for Automatic Evaluation of Machine Translation. ACL 2002

    2002

  2. [10]

    Maja Popović. 2015. chrF: Character n-gram F-score for Automatic MT Evaluation. WMT 2015

    2015

  3. [11]

    Matt Post. 2018. A Call for Clarity in Reporting BLEU Scores. WMT 2018

    2018

  4. [12]

    Philipp Koehn and Rebecca Knowles. 2017. Six Challenges for Neural Machine Translation. ACL Workshop on NMT

    2017

  5. [13]

    Gongbo Tang, Rico Sennrich, and Joakim Nivre. 2018. An Analysis of Attention Mechanisms: The Case of Word Sense Disambiguation in Neural Machine Translation. WMT 2018

    2018

  6. [14]

    Hamidreza Ghader and Christof Monz. 2017. What Does Attention in Neural Machine Translation Pay Attention To? IJCNLP 2017

    2017

  7. [15]

    Jasmijn Bastings, Wilker Aziz, Ivan Titov, and Khalil Sima’an. 2017. Graph Convolutional Encoders for Syntax-Aware Neural Machine Translation. EMNLP 2017

    2017

  8. [16]

    Khapra, and Pushpak Bhattacharyya

    Gowtham Ramesh, Sumanth Doddapaneni, Rahul Aralikatte, Anoop Kunchukuttan, Mitesh M. Khapra, and Pushpak Bhattacharyya. 2022. IndicTrans: A Multilingual Transformer-Based Machine Translation Model for Indian Languages. ACL 2022

    2022

  9. [17]

    Rahul Raja and Arpita Vats. 2025. Parallel Corpora for Machine Translation in Low-Resource Indic Languages: A Comprehensive Review

    2025

  10. [18]

    Sarubi Thillainathan et al. 2025. Beyond Vanilla Fine-Tuning: Leveraging Multistage, Multilingual, and Domain-Specific Methods for Low-Resource Machine Translation

    2025

  11. [19]

    Mikel Artetxe, Sebastian Ruder, and Dani Yogatama. 2020. On the Cross-Lingual Transferability of Monolingual Representations

    2020

  12. [20]

    Roee Aharoni, Melvin Johnson, and Orhan Firat. 2019. Massively Multilingual Neural Machine Translation. NAACL 2019

    2019

  13. [21]

    Melvin Johnson et al. 2017. Google’s Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation. TACL

    2017

  14. [22]

    Surafel Melaku Lakew, Marcello Federico, Matteo Negri, and Marco Turchi. 2018. A Survey of Low- Resource Neural Machine Translation

    2018

  15. [23]

    Vandan Mujadia, Ashok Urlana, Yash Bhaskar, et al. 2024. Assessing Translation Capabilities of Large Language Models Involving English and Indian Languages. EAMT 2024

    2024

  16. [24]

    Marwan Hendy et al. 2023. How Good Are GPT Models at Machine Translation? A Comprehensive Evaluation

    2023

  17. [25]

    Wenxiang Jiao et al. 2023. Is ChatGPT A Good Translator? A Preliminary Study

    2023

Pith tools

Reviewed July 31, 2026 · model on record in the stance chip above.