Pith. sign in

REVIEW 15 cited by

A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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 2309.11674 v2 pith:YZOO6BG7 submitted 2023-09-20 cs.CL

Haoran Xu , Young Jin Kim , Amr Sharaf , Hany Hassan Awadalla This is my paper · ORCID
classification cs.CL
keywords translationfine-tuningmodelmodelsdatadirectionslanguagellms
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Generative Large Language Models (LLMs) have achieved remarkable advancements in various NLP tasks. However, these advances have not been reflected in the translation task, especially those with moderate model sizes (i.e., 7B or 13B parameters), which still lag behind conventional supervised encoder-decoder translation models. Previous studies have attempted to improve the translation capabilities of these moderate LLMs, but their gains have been limited. In this study, we propose a novel fine-tuning approach for LLMs that is specifically designed for the translation task, eliminating the need for the abundant parallel data that traditional translation models usually depend on. Our approach consists of two fine-tuning stages: initial fine-tuning on monolingual data followed by subsequent fine-tuning on a small set of high-quality parallel data. We introduce the LLM developed through this strategy as Advanced Language Model-based trAnslator (ALMA). Based on LLaMA-2 as our underlying model, our results show that the model can achieve an average improvement of more than 12 BLEU and 12 COMET over its zero-shot performance across 10 translation directions from the WMT'21 (2 directions) and WMT'22 (8 directions) test datasets. The performance is significantly better than all prior work and even superior to the NLLB-54B model and GPT-3.5-text-davinci-003, with only 7B or 13B parameters. This method establishes the foundation for a novel training paradigm in machine translation.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 15 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. ConsistencyChecker: Tree-based Evaluation of LLM Generalization Capabilities

    cs.AI 2025-06 conditional novelty 6.0 of 10

    ConsistencyChecker ranks LLMs by how well they survive chains of reversible transformations, and those scores track WMT 2024 translation quality rankings (r > 0.7) without using WMT paired data.

  2. Calibrating Translation Decoding with Quality Estimation on LLMs

    cs.CL 2025-04 conditional novelty 6.0 of 10

    Optimizing the Pearson correlation between hypothesis likelihood and an external quality score during fine-tuning improves LLM translation quality and turns log-likelihood into a competitive reference-free quality estimator.

  3. CRPO: Confidence-Reward Driven Preference Optimization for Machine Translation

    cs.CL 2025-01 conditional novelty 6.0 of 10

    A confidence-reward score for selecting preference pairs improves DPO-based machine translation fine-tuning over reward-only selection methods on ALMA-7B and NLLB-1.3B.

  4. MageBench: Bridging Large Multimodal Models to Agents

    cs.CV 2024-12 conditional novelty 6.0 of 10

    MageBench introduces a 483-scenario benchmark showing current large multimodal models are far weaker than humans at agent tasks requiring continuous visual feedback and planning.

  5. $M^2PO$: Multi-Perspective Multi-Pair Preference Optimization for Machine Translation

    cs.CL 2025-10 reject novelty 5.0 of 10

    M2PO combines a QE-plus-alignment reward with a dynamic curriculum and multi-pair DPO loss, and reports WMT21-22 gains for a 7B translation model, but the abstract's WMT23/24 9B parity claims are unsupported.

  6. A Framework for Generating Conversational Recommendation Datasets from Behavioral Interactions

    cs.IR 2025-06 reject novelty 5.0 of 10

    ConvRecStudio generates roughly 38K synthetic multi-turn recommendation dialogs across three domains from historical interactions, and a cross-attention transformer fusing history with dialog beats dialog-only and his...

  7. RIVAL: Reinforcement Learning with Iterative and Adversarial Optimization for Machine Translation

    cs.CL 2025-06 conditional novelty 5.0 of 10

    RIVAL iteratively re-trains a reward model adversarially against the current translator and adds a BLEU-predicting head, improving in-domain WMT and subtitle translation over SFT baselines.

  8. ConECT Dataset: Overcoming Data Scarcity in Context-Aware E-Commerce MT

    cs.CL 2025-06 conditional novelty 5.0 of 10

    A new Czech-to-Polish e-commerce translation dataset is released, and the paper shows small improvements from visual and category context, with a negative result for image descriptions.

  9. Mutarjim: Advancing Bidirectional Arabic-English Translation with a Small Language Model

    cs.CL 2025-05 reject novelty 5.0 of 10

    A compact 1.5B Arabic-English model beats GPT-4o mini only on the authors' own Tarjama-25 benchmark, while trailing large models on standard WMT24++ and IWSLT2017 tests.

  10. Multilingual Contextualization of Large Language Models for Document-Level Machine Translation

    cs.CL 2025-04 conditional novelty 5.0 of 10

    Fine-tuning translation LLMs on a curated multi-domain document dataset (DocBlocks) with document, chunk, and sentence level instruction formats improves document-level translation quality and speed while preserving s...

  11. Self-Evolution Knowledge Distillation for LLM-based Machine Translation

    cs.CL 2024-12 conditional novelty 5.0 of 10

    A token-adaptive distillation method that mixes teacher and ground-truth targets only for hard tokens yields consistent BLEU gains in LLM translation.

  12. CHAI for LLMs: Improving Code-Mixed Translation in Large Language Models through Reinforcement Learning with AI Feedback

    cs.CL 2024-11 conditional novelty 5.0 of 10

    CHAI trains a reward model on GPT-4o preference labels and uses PPO to align Llama-3.1-8B for English-to-Hinglish translation, claiming a 25.66% human win-rate improvement over baselines.

  13. Beyond the Sentence: A Survey on Context-Aware Machine Translation with Large Language Models

    cs.CL 2025-06 conditional novelty 4.0 of 10

    A survey of context-aware machine translation with large language models, categorizing prompting, fine-tuning, and agent-based approaches.

  14. A comparison of translation performance between DeepL and Supertext

    cs.CL 2025-02 conditional novelty 4.0 of 10

    In a blind professional A/B evaluation with full document context, Supertext is preferred at document level in three of four language directions, while segment-level preferences are mostly tied.

  15. Domain-Specific Translation with Open-Source Large Language Models: Resource-Oriented Analysis

    cs.CL 2024-12 conditional novelty 4.0 of 10

    For medical translation, NLLB-200 3.3B matches or beats 7-8B open-source LLMs in most directions, and fine-tuning it is often as good as fine-tuning much larger LLMs.

Pith tools