Reuse Your Rewards: Reward Model Transfer for Zero-Shot Cross-Lingual Alignment

• URL: http://arxiv.org/abs/2404.12318v2
• Date: Mon, 14 Oct 2024 17:58:05 GMT
• Title: Reuse Your Rewards: Reward Model Transfer for Zero-Shot Cross-Lingual Alignment
• Authors: Zhaofeng Wu, Ananth Balashankar, Yoon Kim, Jacob Eisenstein, Ahmad Beirami,
• Abstract summary: We evaluate a simple approach for zero-shot cross-lingual alignment. Cross-lingually aligned models are preferred by humans over unaligned models. A different-language reward model sometimes yields better aligned models than a same-language reward model.
• Score: 39.94156255629528
• License:
• Abstract: Aligning language models (LMs) based on human-annotated preference data is a crucial step in obtaining practical and performant LM-based systems. However, multilingual human preference data are difficult to obtain at scale, making it challenging to extend this framework to diverse languages. In this work, we evaluate a simple approach for zero-shot cross-lingual alignment, where a reward model is trained on preference data in one source language and directly applied to other target languages. On summarization and open-ended dialog generation, we show that this method is consistently successful under comprehensive evaluation settings, including human evaluation: cross-lingually aligned models are preferred by humans over unaligned models on up to >70% of evaluation instances. We moreover find that a different-language reward model sometimes yields better aligned models than a same-language reward model. We also identify best practices when there is no language-specific data for even supervised finetuning, another component in alignment.

Related papers

• Investigating on RLHF methodology [0.0]
  We discuss the features of training a Preference Model, which simulates human preferences, and the methods and details we found essential for achieving the best results. We also discuss using Reinforcement Learning to fine-tune Large Language Models and describe the challenges we faced and the ways to overcome them.
  arXiv  Detail & Related papers  (2024-10-02T17:46:22Z)
• An Open Dataset and Model for Language Identification [84.15194457400253]
  We present a LID model which achieves a macro-average F1 score of 0.93 and a false positive rate of 0.033 across 201 languages. We make both the model and the dataset available to the research community.
  arXiv  Detail & Related papers  (2023-05-23T08:43:42Z)
• Cross-language Sentence Selection via Data Augmentation and Rationale Training [22.106577427237635]
  It uses data augmentation and negative sampling techniques on noisy parallel sentence data to learn a cross-lingual embedding-based query relevance model. Results show that this approach performs as well as or better than multiple state-of-the-art machine translation + monolingual retrieval systems trained on the same parallel data.
  arXiv  Detail & Related papers  (2021-06-04T07:08:47Z)
• Unsupervised Domain Adaptation of a Pretrained Cross-Lingual Language Model [58.27176041092891]
  Recent research indicates that pretraining cross-lingual language models on large-scale unlabeled texts yields significant performance improvements. We propose a novel unsupervised feature decomposition method that can automatically extract domain-specific features from the entangled pretrained cross-lingual representations. Our proposed model leverages mutual information estimation to decompose the representations computed by a cross-lingual model into domain-invariant and domain-specific parts.
  arXiv  Detail & Related papers  (2020-11-23T16:00:42Z)
• Towards Zero-Shot Multilingual Synthetic Question and Answer Generation for Cross-Lingual Reading Comprehension [20.570539023748424]
  We propose a simple method to generate multilingual question and answer pairs on a large scale. These synthetic samples can be used to improve the zero-shot performance of multilingual QA models on target languages.
  arXiv  Detail & Related papers  (2020-10-22T19:59:37Z)
• Comparison of Interactive Knowledge Base Spelling Correction Models for Low-Resource Languages [81.90356787324481]
  Spelling normalization for low resource languages is a challenging task because the patterns are hard to predict. This work shows a comparison of a neural model and character language models with varying amounts on target language data. Our usage scenario is interactive correction with nearly zero amounts of training examples, improving models as more data is collected.
  arXiv  Detail & Related papers  (2020-10-20T17:31:07Z)
• Mixed-Lingual Pre-training for Cross-lingual Summarization [54.4823498438831]
  Cross-lingual Summarization aims at producing a summary in the target language for an article in the source language. We propose a solution based on mixed-lingual pre-training that leverages both cross-lingual tasks like translation and monolingual tasks like masked language models. Our model achieves an improvement of 2.82 (English to Chinese) and 1.15 (Chinese to English) ROUGE-1 scores over state-of-the-art results.
  arXiv  Detail & Related papers  (2020-10-18T00:21:53Z)
• Model Selection for Cross-Lingual Transfer [15.197350103781739]
  We propose a machine learning approach to model selection that uses the fine-tuned model's own internal representations to predict its cross-lingual capabilities. In extensive experiments we find that this method consistently selects better models than English validation data across twenty five languages.
  arXiv  Detail & Related papers  (2020-10-13T02:36:48Z)
• Cross-lingual Spoken Language Understanding with Regularized Representation Alignment [71.53159402053392]
  We propose a regularization approach to align word-level and sentence-level representations across languages without any external resource. Experiments on the cross-lingual spoken language understanding task show that our model outperforms current state-of-the-art methods in both few-shot and zero-shot scenarios.
  arXiv  Detail & Related papers  (2020-09-30T08:56:53Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.