Alignment Techniques and Intermediate-Task Training for Zero-Shot Cross-Lingual Transfer in XTREME-R

Transfer learning from large language models (LLMs) has emerged as a powerful technique to enable knowledge-based fine-tuning for a number of tasks, adaptation of models for different domains and even languages. However, it remains an open question, if and when transfer learning will work, i.e. leading to positive or negative transfer. In this paper, we analyze the knowledge transfer across three natural language processing (NLP) tasks - text classification, sentimental analysis, and sentence similarity, using three LLMs - BERT, RoBERTa, and XLNet - and analyzing their performance, by fine-tun

Research goal: Can alignment techniques (e.g., fine-tuning with human feedback) further improve zero-shot cross-lingual transfer performance in XTREME-R when combined with intermediate-task training, as measured by accuracy and reasoning scores?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.9/10.