Zihao Li, Jörg Tiedemann, Shaoxiong Ji, Timothee Mickus, Vincent Segonne
We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Pretrained language models (PLMs) display impressive performances and have captured the attention of the NLP community. Establishing best practices in pretraining has, therefore, become a major focus of NLP research, especially since insights gained from monolingual English models may not necessarily apply to more complex multilingual models. One significant caveat of the current state of the art is that different works are rarely comparable: they often discuss different parameter counts, training data, and evaluation methodology. This paper proposes a comparison of multilingual pretraining objectives in a controlled methodological environment. We ensure that training data and model architectures are comparable, and discuss the downstream performances across 6 languages that we observe in probing and fine-tuning scenarios. We make two key observations: (1) the architecture dictates which pretraining objective is optimal; (2) multilingual translation is a very effective pretraining objective under the right conditions. We make our code, data, and model weights available at https://github. com/Helsinki-NLP/lm-vs-mt.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2407.15489")
get_code_for_paper("2407.15489")
have("2407.15489")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2407.15489.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2407.15489)
Connect an agent — have() is free.