# Better Alignment with Instruction Back-and-Forth Translation

arXiv: https://arxiv.org/abs/2408.04614 · HTML: https://mlsanity.com/paper/2408.04614/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- Placed in LLM preference optimization methods, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 25 of its 40 references that are in the map.
- First posted on arXiv 8 Aug 2024.

## Placed in

- LLM alignment / RLHF / preference optimisation / [LLM preference optimization methods](https://mlsanity.com/alignment/llm-preference-optimization-methods/)

## Builds on (most cited first, 25 of 25)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [Self-Instruct: Aligning Language Models with Self-Generated Instructions](https://mlsanity.com/paper/2212.10560/) (2022)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [WizardLM: Empowering large pre-trained language models to follow complex instructions](https://mlsanity.com/paper/2304.12244/) (2023)
- [LIMA: Less Is More for Alignment](https://mlsanity.com/paper/2305.11206/) (2023)
- [Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluators](https://mlsanity.com/paper/2404.04475/) (2024)
- [Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Tasks](https://mlsanity.com/paper/2204.07705/) (2022)
- [The Flan Collection: Designing Data and Methods for Effective Instruction Tuning](https://mlsanity.com/paper/2301.13688/) (2023)
- [OpenAssistant Conversations -- Democratizing Large Language Model Alignment](https://mlsanity.com/paper/2304.07327/) (2023)
- [Cross-Task Generalization via Natural Language Crowdsourcing Instructions](https://mlsanity.com/paper/2104.08773/) (2021)
- [Instruction Tuning for Large Language Models: A Survey](https://mlsanity.com/paper/2308.10792/) (2023)
- [Zephyr: Direct Distillation of LM Alignment](https://mlsanity.com/paper/2310.16944/) (2023)
- [MAUVE: Measuring the Gap Between Neural Text and Human Text using Divergence Frontiers](https://mlsanity.com/paper/2102.01454/) (2021)
- [How Far Can Camels Go? Exploring the State of Instruction Tuning on Open Resources](https://mlsanity.com/paper/2306.04751/) (2023)
- [What Makes Good Data for Alignment? A Comprehensive Study of Automatic Data Selection in Instruction Tuning](https://mlsanity.com/paper/2312.15685/) (2023)
- [The False Promise of Imitating Proprietary LLMs](https://mlsanity.com/paper/2305.15717/) (2023)
- [Rephrasing the Web: A Recipe for Compute and Data-Efficient Language Modeling](https://mlsanity.com/paper/2401.16380/) (2024)
- [#InsTag: Instruction Tagging for Analyzing Supervised Fine-tuning of Large Language Models](https://mlsanity.com/paper/2308.07074/) (2023)
- [Long Is More for Alignment: A Simple but Tough-to-Beat Baseline for Instruction Fine-Tuning](https://mlsanity.com/paper/2402.04833/) (2024)

## Built on by (most cited first, 16 of 16)

- [Self-Boosting Large Language Models with Synthetic Preference Data](https://mlsanity.com/paper/2410.06961/) (2024)
- [Magnet: Multi-turn Tool-use Data Synthesis and Distillation via Graph Translation](https://mlsanity.com/paper/2503.07826/) (2025)
- [Source2Synth: Synthetic Data Generation and Curation Grounded in Real Data Sources](https://mlsanity.com/paper/2409.08239/) (2024)
- [AIR: Complex Instruction Generation via Automatic Iterative Refinement](https://mlsanity.com/paper/2502.17787/) (2025)
- [Skill Self-Play: Pushing the Frontier of LLM Capability with Co-Evolving Skills](https://mlsanity.com/paper/2607.22529/) (2026)
- [Aligning Large Language Models with Implicit Preferences from User-Generated Content](https://mlsanity.com/paper/2506.04463/) (2025)
- [Instruction-Tuning Data Synthesis from Scratch via Web Reconstruction](https://mlsanity.com/paper/2504.15573/) (2025)
- [IFDECORATOR: Wrapping Instruction Following Reinforcement Learning with Verifiable Rewards](https://mlsanity.com/paper/2508.04632/) (2025)
- [UPDESH: Synthesizing Grounded Instruction Tuning Data for 13 Indic Languages](https://mlsanity.com/paper/2509.21294/) (2025)
- [Backtranslation and paraphrasing in the LLM era? Comparing data augmentation methods for emotion classification](https://mlsanity.com/paper/2507.14590/) (2025)
- [Small Reward Models via Backward Inference](https://mlsanity.com/paper/2602.13551/) (2026)
- [Large Language Models for Mental Health: A Multilingual Evaluation](https://mlsanity.com/paper/2602.02440/) (2026)
- [MAIN: Mutual Alignment Is Necessary for instruction tuning](https://mlsanity.com/paper/2504.12913/) (2025)
- [Merge and Conquer: Instructing Multilingual Models by Adding Target Language Weights](https://mlsanity.com/paper/2603.28263/) (2026)
- [Synthesizing Instruction-Tuning Datasets with Contrastive Decoding](https://mlsanity.com/paper/2604.13538/) (2026)
- [UNSPECIFIC: General Constraint Synthesis for Breaking Copy-and-Paste Shortcut in LLM Instruction Following](https://mlsanity.com/paper/2608.09154/) (2026)

## Abstract (arXiv)

We propose a new method, instruction back-and-forth translation, to construct high-quality synthetic data grounded in world knowledge for aligning large language models (LLMs). Given documents from a web corpus, we generate and curate synthetic instructions using the backtranslation approach proposed by Li et al.(2023a), and rewrite the responses to improve their quality further based on the initial documents. Fine-tuning with the resulting (backtranslated instruction, rewritten response) pairs yields higher win rates on AlpacaEval than using other common instruction datasets such as Humpback, ShareGPT, Open Orca, Alpaca-GPT4 and Self-instruct. We also demonstrate that rewriting the responses with an LLM outperforms direct distillation, and the two generated text distributions exhibit significant distinction in embedding space. Further analysis shows that our backtranslated instructions are of higher quality than other sources of synthetic instructions, while our responses are more diverse and complex than those obtained from distillation. Overall we find that instruction back-and-forth translation combines the best of both worlds -- making use of the information diversity and quantity found on the web, while ensuring the quality of the responses which is necessary for effective alignment.