# Boosting LLM Translation Skills without General Ability Loss via Rationale Distillation

arXiv: https://arxiv.org/abs/2410.13944 · HTML: https://mlsanity.com/paper/2410.13944/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 37 of its 56 references that are in the map.
- First posted on arXiv 17 Oct 2024.

## Builds on (most cited first, 30 of 37)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://mlsanity.com/paper/2403.13372/) (2024)
- [Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluators](https://mlsanity.com/paper/2404.04475/) (2024)
- [Sequence-Level Knowledge Distillation](https://mlsanity.com/paper/1606.07947/) (2016)
- [Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes](https://mlsanity.com/paper/2305.02301/) (2023)
- [The FLORES-101 Evaluation Benchmark for Low-Resource and Multilingual Machine Translation](https://mlsanity.com/paper/2106.03193/) (2021)
- [LESS: Selecting Influential Data for Targeted Instruction Tuning](https://mlsanity.com/paper/2402.04333/) (2024)
- [Magicoder: Empowering Code Generation with OSS-Instruct](https://mlsanity.com/paper/2312.02120/) (2023)
- [Specializing Smaller Language Models towards Multi-Step Reasoning](https://mlsanity.com/paper/2301.12726/) (2023)
- [Continual Learning of Large Language Models: A Comprehensive Survey](https://mlsanity.com/paper/2404.16789/) (2024)
- [Fine-tuned Language Models are Continual Learners](https://mlsanity.com/paper/2205.12393/) (2022)
- [Continual Learning for Large Language Models: A Survey](https://mlsanity.com/paper/2402.01364/) (2024)
- [SCOTT: Self-Consistent Chain-of-Thought Distillation](https://mlsanity.com/paper/2305.01879/) (2023)
- [Self-Distillation Bridges Distribution Gap in Language Model Fine-Tuning](https://mlsanity.com/paper/2402.13669/) (2024)
- [A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models](https://mlsanity.com/paper/2309.11674/) (2023)
- [Continual Learning of Natural Language Processing Tasks: A Survey](https://mlsanity.com/paper/2211.12701/) (2022)
- [InsCL: A Data-efficient Continual Learning Paradigm for Fine-tuning Large Language Models with Instructions](https://mlsanity.com/paper/2403.11435/) (2024)
- [LLaMAX: Scaling Linguistic Horizons of LLM by Enhancing Translation Capabilities Beyond 100 Languages](https://mlsanity.com/paper/2407.05975/) (2024)
- [Fine-tuning Large Language Models for Domain-specific Machine Translation](https://mlsanity.com/paper/2402.15061/) (2024)
- [BayLing: Bridging Cross-lingual Alignment and Instruction Following through Interactive Translation for Large Language Models](https://mlsanity.com/paper/2306.10968/) (2023)
- [ParroT: Translating during Chat using Large Language Models tuned with Human Translation and Feedback](https://mlsanity.com/paper/2304.02426/) (2023)
- [ConTinTin: Continual Learning from Task Instructions](https://mlsanity.com/paper/2203.08512/) (2022)
- [CoCoA-MT: A Dataset and Benchmark for Contrastive Controlled MT with Application to Formality](https://mlsanity.com/paper/2205.04022/) (2022)
- [The Fine-Tuning Paradox: Boosting Translation Quality Without Sacrificing LLM Abilities](https://mlsanity.com/paper/2405.20089/) (2024)

## Built on by (most cited first, 6 of 6)

- [Improving MLLM's Document Image Machine Translation via Synchronously Self-reviewing Its OCR Proficiency](https://mlsanity.com/paper/2507.08309/) (2025)
- [KD4MT: A Survey of Knowledge Distillation for Machine Translation](https://mlsanity.com/paper/2602.15845/) (2026)
- [GemMaroc: Unlocking Darija Proficiency in LLMs with Minimal Data](https://mlsanity.com/paper/2505.17082/) (2025)
- [Can Large Language Models Anticipate Behavioral Responses to Social Policies? A Case of Pension Enrollment Prediction among China's Flexible Workers](https://mlsanity.com/paper/2609.05189/) (2026)
- [Stemming Hallucination in Language Models Using a Licensing Oracle](https://mlsanity.com/paper/2511.06073/) (2025)
- [Audit Before You Distill: A Controlled Workflow for Reasoning Transfer into Compact Medical Language Models](https://mlsanity.com/paper/2610.16713/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) have achieved impressive results across numerous NLP tasks but still encounter difficulties in machine translation. Traditional methods to improve translation have typically involved fine-tuning LLMs using parallel corpora. However, vanilla fine-tuning often leads to catastrophic forgetting of the instruction-following capabilities and alignment with human preferences, compromising their broad general abilities and introducing potential security risks. These abilities, which are developed using proprietary and unavailable training data, make existing continual instruction tuning methods ineffective. To overcome this issue, we propose a novel approach called RaDis (Rationale Distillation). RaDis harnesses the strong generative capabilities of LLMs to create rationales for training data, which are then "replayed" to prevent forgetting. These rationales encapsulate general knowledge and safety principles, acting as self-distillation targets to regulate the training process. By jointly training on both reference translations and self-generated rationales, the model can learn new translation skills while preserving its overall general abilities. Extensive experiments demonstrate that our method enhances machine translation performance while maintaining the broader capabilities of LLMs across other tasks. This work presents a pathway for creating more versatile LLMs that excel in specialized tasks without compromising generality and safety.