AlexaTM 20B: Few-Shot Learning Using a Large-Scale Multilingual Seq2Seq Model

• Published in: arXiv.org 2022-08
• Main Authors: Soltan, Saleh, Ananthakrishnan, Shankar, FitzGerald, Jack, Gupta, Rahul, Hamza, Wael, Khan, Haidar, Peris, Charith, Rawls, Stephen, Rosenbaum, Andy, Rumshisky, Anna, Prakash, Chandana Satya, Sridhar, Mukund, Triefenbach, Fabian, Verma, Apurv, Tur, Gokhan, Natarajan, Prem
• Format: Article
• Language: English
• ISSN: 2331-8422

In this work, we demonstrate that multilingual large-scale sequence-to-sequence (seq2seq) models, pre-trained on a mixture of denoising and Causal Language Modeling (CLM) tasks, are more efficient few-shot learners than decoder-only models on various tasks. In particular, we train a 20 billion parameter multilingual seq2seq model called Alexa Teacher Model (AlexaTM 20B) and show that it achieves state-of-the-art (SOTA) performance on 1-shot summarization tasks, outperforming a much larger 540B PaLM decoder model. AlexaTM 20B also achieves SOTA in 1-shot machine translation, especially for low-resource languages, across almost all language pairs supported by the model (Arabic, English, French, German, Hindi, Italian, Japanese, Marathi, Portuguese, Spanish, Tamil, and Telugu) on Flores-101 dataset. We also show in zero-shot setting, AlexaTM 20B outperforms GPT3 (175B) on SuperGLUE and SQuADv2 datasets and provides SOTA performance on multilingual tasks such as XNLI, XCOPA, Paws-X, and XWinograd. Overall, our results present a compelling case for seq2seq models as a powerful alternative to decoder-only models for Large-scale Language Model (LLM) training.