# Prefix Text as a Yarn: Eliciting Non-English Alignment in Foundation Language Model

arXiv: https://arxiv.org/abs/2404.16766 · HTML: https://mlsanity.com/paper/2404.16766/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 52 references that are in the map.
- First posted on arXiv 25 Apr 2024.

## Builds on (most cited first, 30 of 34)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [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)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Neural Machine Translation by Jointly Learning to Align and Translate](https://mlsanity.com/paper/1409.0473/) (2014)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing](https://mlsanity.com/paper/1808.06226/) (2018)
- [No Language Left Behind: Scaling Human-Centered Machine Translation](https://mlsanity.com/paper/2207.04672/) (2022)
- [LIMA: Less Is More for Alignment](https://mlsanity.com/paper/2305.11206/) (2023)
- [COMET: A Neural Framework for MT Evaluation](https://mlsanity.com/paper/2009.09025/) (2020)
- [A Survey on In-context Learning](https://mlsanity.com/paper/2301.00234/) (2022)
- [RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment](https://mlsanity.com/paper/2304.06767/) (2023)
- [An Empirical Study of Catastrophic Forgetting in Large Language Models During Continual Fine-tuning](https://mlsanity.com/paper/2308.08747/) (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)
- [Word Translation Without Parallel Data](https://mlsanity.com/paper/1710.04087/) (2017)
- [Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision](https://mlsanity.com/paper/2312.09390/) (2023)
- [What Makes Good Data for Alignment? A Comprehensive Study of Automatic Data Selection in Instruction Tuning](https://mlsanity.com/paper/2312.15685/) (2023)
- [Principle-Driven Self-Alignment of Language Models from Scratch with Minimal Human Supervision](https://mlsanity.com/paper/2305.03047/) (2023)
- [The Unlocking Spell on Base LLMs: Rethinking Alignment via In-Context Learning](https://mlsanity.com/paper/2312.01552/) (2023)
- [How Abilities in Large Language Models are Affected by Supervised Fine-tuning Data Composition](https://mlsanity.com/paper/2310.05492/) (2023)
- [XGLUE: A New Benchmark Dataset for Cross-lingual Pre-training, Understanding and Generation](https://mlsanity.com/paper/2004.01401/) (2020)
- [RAIN: Your Language Models Can Align Themselves without Finetuning](https://mlsanity.com/paper/2309.07124/) (2023)
- [Unicoder: A Universal Language Encoder by Pre-training with Multiple Cross-lingual Tasks](https://mlsanity.com/paper/1909.00964/) (2019)
- [MM-LIMA: Less Is More for Alignment in Multi-Modal Datasets](https://mlsanity.com/paper/2308.12067/) (2023)
- [Investigating the Translation Performance of a Large Multilingual Language Model: the Case of BLOOM](https://mlsanity.com/paper/2303.01911/) (2023)
- [Selective Reflection-Tuning: Student-Selected Data Recycling for LLM Instruction-Tuning](https://mlsanity.com/paper/2402.10110/) (2024)
- [Active Instruction Tuning: Improving Cross-Task Generalization by Training on Prompt Sensitive Tasks](https://mlsanity.com/paper/2311.00288/) (2023)
- [xCoT: Cross-lingual Instruction Tuning for Cross-lingual Chain-of-Thought Reasoning](https://mlsanity.com/paper/2401.07037/) (2024)

## Built on by (most cited first, 10 of 10)

- [Understanding Aha Moments: from External Observations to Internal Mechanisms](https://mlsanity.com/paper/2504.02956/) (2025)
- [DetectRL: Benchmarking LLM-Generated Text Detection in Real-World Scenarios](https://mlsanity.com/paper/2410.23746/) (2024)
- [Who Wrote This? The Key to Zero-Shot LLM-Generated Text Detection Is GECScore](https://mlsanity.com/paper/2405.04286/) (2024)
- [What is the Best Way for ChatGPT to Translate Poetry?](https://mlsanity.com/paper/2406.03450/) (2024)
- [Not All LoRA Parameters Are Essential: Insights on Inference Necessity](https://mlsanity.com/paper/2503.23360/) (2025)
- [FOCUS: Forging Originality through Contrastive Use in Self-Plagiarism for Language Models](https://mlsanity.com/paper/2406.00839/) (2024)
- [Intrinsic Model Weaknesses: How Priming Attacks Unveil Vulnerabilities in Large Language Models](https://mlsanity.com/paper/2502.16491/) (2025)
- [Take Off the Training Wheels Progressive In-Context Learning for Effective Alignment](https://mlsanity.com/paper/2503.09958/) (2025)
- [A Survey on Training-free Alignment of Large Language Models](https://mlsanity.com/paper/2508.09016/) (2025)
- [Multilinguality at the Edge: Developing Language Models for the Global South](https://mlsanity.com/paper/2604.21637/) (2026)

## Abstract (arXiv)

While supervised fine-tuning (SFT) has been a straightforward approach for tailoring the output of foundation large language model (LLM) to specific preferences, concerns have been raised about the depth of this alignment, with some critiques suggesting it is merely "superficial". We critically examine this hypothesis within the scope of cross-lingual generation tasks, proposing that the effectiveness of SFT may be constrained by its reliance on prior tokens to guide cross-lingual generation. Based on this crucial insight, and in response to the challenges posed by the costly and limited availability of non-English data for SFT, we introduce a novel training-free alignment method named PreTTY, which employs minimal task-related prior tokens to bridge the foundation LLM and the SFT LLM, achieving comparable performance without training. Experiments on machine translation and part-of-speech tagging across eight languages demonstrate the efficacy of PreTTY in cross-lingual settings. Remarkably, by initiating the decoding process with only one or two prior tokens, foundation LLMs can achieve performance comparable to their SFT counterparts. This method presents a cost-effective alternative to SFT and advances the democratization of multilingual LLMs.