Pith. sign in

REVIEW 3 cited by

ChatGPT MT: Competitive for High- (but not Low-) Resource Languages

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2309.07423 v1 pith:FRFUXLBV submitted 2023-09-14 cs.CL

classification cs.CL
keywords languageschatgptlanguagellmsanalysisevidenceexperimentallrls
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Large language models (LLMs) implicitly learn to perform a range of language tasks, including machine translation (MT). Previous studies explore aspects of LLMs' MT capabilities. However, there exist a wide variety of languages for which recent LLM MT performance has never before been evaluated. Without published experimental evidence on the matter, it is difficult for speakers of the world's diverse languages to know how and whether they can use LLMs for their languages. We present the first experimental evidence for an expansive set of 204 languages, along with MT cost analysis, using the FLORES-200 benchmark. Trends reveal that GPT models approach or exceed traditional MT model performance for some high-resource languages (HRLs) but consistently lag for low-resource languages (LRLs), under-performing traditional MT for 84.1% of languages we covered. Our analysis reveals that a language's resource level is the most important feature in determining ChatGPT's relative ability to translate it, and suggests that ChatGPT is especially disadvantaged for LRLs and African languages.

Discussion (0). Sign in to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 5 citations worldwide. Full citation record

  1. Dutch CrowS-Pairs: Adapting a Challenge Dataset for Measuring Social Biases in Language Models for Dutch

    cs.CL 2025-07 conditional novelty 6.0 of 10

    The paper presents a Dutch adaptation of the CrowS-Pairs bias benchmark and reports bias scores for seven masked and two autoregressive language models across nine demographic categories.

  2. Toxicity-Aware Few-Shot Prompting for Low-Resource Singlish Translation

    cs.CL 2025-07 conditional novelty 4.0 of 10

    A two-stage pipeline combining human-curated Singlish examples with embedding-based LLM ranking selects GPT-4o mini for toxicity-preserving translation, reaching gold-level human scores for Chinese and Malay but not Tamil.

  3. Bridging the Gap with Retrieval-Augmented Generation: Making Prosthetic Device User Manuals Available in Marginalised Languages

    cs.LG 2025-06 reject novelty 2.0 of 10

    A proposed RAG-based framework for translating prosthetic device manuals into marginalised languages is described, but no results are presented.

Pith tools