One For All & All For One: Bypassing Hyperparameter Tuning with Model Averaging For Cross-Lingual Transfer
Multilingual language models enable zero-shot cross-lingual transfer (ZS-XLT): fine-tuned on sizable source-language task data, they perform the task in target languages without labeled instances. The effectiveness of ZS-XLT hinges on the linguistic proximity between languages and the amount of pretraining data for a language. Because of this, model selection based on source-language validation is unreliable: it picks model snapshots with suboptimal target-language performance. As a remedy, some work optimizes ZS-XLT by extensively tuning hyperparameters: the follow-up work then routinely struggles to replicate the original results. Other work searches over narrower hyperparameter grids, reporting substantially lower performance. In this work, we therefore propose an unsupervised evaluation protocol for ZS-XLT that decouples performance maximization from hyperparameter tuning. As a robust and more transparent alternative to extensive hyperparameter tuning, we propose to accumulatively average snapshots from different runs into a single model. We run broad ZS-XLT experiments on both higher-level semantic tasks (NLI, extractive QA) and a lower-level token classification task (NER) and find that conventional model selection based on source-language validation quickly plateaus to suboptimal ZS-XLT performance. On the other hand, our accumulative run-by-run averaging of models trained with different hyperparameters boosts ZS-XLT performance and closely correlates with "oracle" ZS-XLT, i.e., model selection based on target-language validation performance.
Code (1)
Tasks
AllCross-Lingual TransferModel SelectionNERtoken-classificationToken ClassificationZero-Shot Cross-Lingual TransferSimilar Papers 제목 키워드 기반
Free Lunch: Robust Cross-Lingual Transfer via Model Checkpoint Averaging
Massively multilingual language models have displayed strong performance in zero-shot (ZS-XLT) and few-shot (FS-XLT) cross-lingual transfer setups, where models fine-tuned on task data in a source language are transferre…
Cross-Lingual TransferModel SelectionNERPOS+2Optimistic Dual Averaging Unifies Modern Optimizers
We introduce SODA, a generalization of Optimistic Dual Averaging, which provides a common perspective on state-of-the-art optimizers like Muon, Lion, AdEMAMix and NAdam, showing that they can all be viewed as optimistic …
Frustratingly Simple Regularization to Improve Zero-shot Cross-lingual Robustness
Large-scale multilingual pretrained encoders, such as mBERT and XLM-R, have demonstrated impressive zero-shot cross-lingual transfer capability across multiple NLP tasks. However, as we show in this paper, these models s…
Cross-Lingual TransferXLM-RZero-Shot Cross-Lingual TransferZero-shot cross-lingual transfer in instruction tuning of large language models
Instruction tuning (IT) is widely used to teach pretrained large language models (LLMs) to follow arbitrary instructions, but is under-studied in multilingual settings. In this work, we conduct a systematic study of zero…
Cross-Lingual TransferInstruction FollowingLanguage ModellingLarge Language Model+1Multilingual Safety Alignment Via Sparse Weight Editing
Large Language Models (LLMs) exhibit significant safety disparities across languages, with low-resource languages (LRLs) often bypassing safety guardrails established for high-resource languages (HRLs) like English. Exis…
Reinforcement Learning