2024/07/15 by Wenhao Zhu, Zhu, Wenhao, Sizhe Liu +9 · 2 citations
Computer Science · #Algorithms and Data Compression #Computation and Language (cs.CL) #FOS: Computer and information sciences #Natural Language Processing Techniques #Speech Recognition and Synthesis
paper · pdf · doi:10.48550/arxiv.2407.10795
openalex publication_date 2024/07/15 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Decoding by contrasting layers (DoLa), is designed to improve the generation quality of large language models (LLMs) by contrasting the prediction probabilities between an early exit output (amateur logits) and the final output (expert logits). However, we find that this approach does not work well on non-English tasks. Inspired by previous interpretability work on language transition during the model's forward pass, we discover that this issue arises from a language mismatch between early exit output and final output. In this work, we propose an improved contrastive decoding algorithm that is effective for diverse languages beyond English. To obtain more helpful amateur logits, we devise two strategies to skip a set of bottom, language-agnostic layers based on our preliminary analysis. Experimental results on multilingual reasoning benchmarks demonstrate that our proposed method outperforms previous contrastive decoding baselines and substantially improves LLM's chain-of-thought reasoning accuracy across 11 languages. The project will be available at: https://github.com/NJUNLP/SkipLayerCD.