CLAILGMay 21, 2025

DeFTX: Denoised Sparse Fine-Tuning for Zero-Shot Cross-Lingual Transfer

arXiv:2505.15090v11 citationsh-index: 22EMNLP
Originality Incremental advance
AI Analysis

This work addresses the problem of scaling language models to low-resource languages for NLP practitioners, though it appears incremental as it builds on prior sparse fine-tuning approaches.

The paper tackles the challenge of zero-shot cross-lingual transfer for low-resource languages by introducing DeFT-X, a method that denoises weight matrices before sparse fine-tuning, resulting in performance that matches or exceeds existing baselines on sentiment classification and natural language inference tasks.

Effective cross-lingual transfer remains a critical challenge in scaling the benefits of large language models from high-resource to low-resource languages. Towards this goal, prior studies have explored many approaches to combine task knowledge from task-specific data in a (high-resource) source language and language knowledge from unlabeled text in a (low-resource) target language. One notable approach proposed composable sparse fine-tuning (SFT) for cross-lingual transfer that learns task-specific and language-specific sparse masks to select a subset of the pretrained model's parameters that are further fine-tuned. These sparse fine-tuned vectors (SFTs) are subsequently composed with the pretrained model to facilitate zero-shot cross-lingual transfer to a task in a target language, using only task-specific data from a source language. These sparse masks for SFTs were identified using a simple magnitude-based pruning. In our work, we introduce DeFT-X, a novel composable SFT approach that denoises the weight matrices of a pretrained model before magnitude pruning using singular value decomposition, thus yielding more robust SFTs. We evaluate DeFT-X on a diverse set of extremely low-resource languages for sentiment classification (NusaX) and natural language inference (AmericasNLI) and demonstrate that it performs at par or outperforms SFT and other prominent cross-lingual transfer baselines.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes