patrickvonplaten/data2vec_vision

Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities

UniLM AI

Pre-trained models across tasks (understanding, generation and translation), languages, and modalities

The family of UniLM AI:

UniLM (v1@NeurIPS'19 | v2@ICML'20 | v3@ACL'21): unified pre-training for language understanding and generation

InfoXLM (v1@NAACL'21 | v2@ACL'21): multilingual/cross-lingual pre-trained models for language understanding and generation

DeltaLM (NEW): encoder-decoder pre-training for language generation and translation by augmenting pretrained multilingual encoders

MiniLM (v1@NeurIPS'20 | v2@ACL'21): small and fast pre-trained models for language understanding and generation

AdaLM (v1@ACL'21): domain, language, and task adaptation of pre-trained models

LayoutLM (v1@KDD'20 | v2@ACL'21): multimodal (text + layout/format + image) pre-training for Document AI (e.g. scanned documents, PDF, etc.)

LayoutXLM (NEW): multimodal (text + layout/format + image) pre-training for multilingual document understanding

BEiT (NEW): BERT Pre-Training of Image Transformers

UniSpeech (v1@ICML'21): Speech Pre-Training for ASR and TTS

s2s-ft: sequence-to-sequence fine-tuning toolkit

XLM-T (NEW): Multilingual NMT w/ pretrained cross-lingual encoders

News

• August 2021: BEiT is on HuggingFace
• [Model Release] July, 2021: BEiT - Towards BERT moment for CV
• [Model Release] June, 2021: LayoutLMv2, LayoutXLM, MiniLMv2, and AdaLM.
• May, 2021: LayoutLMv2, InfoXLMv2, MiniLMv2, UniLMv3, and AdaLM were accepted by ACL 2021.
• April, 2021: LayoutXLM is coming by extending the LayoutLM into multilingual support! A multilingual form understanding benchmark XFUND is also introduced, which includes forms with human labeled key-value pairs in 7 languages (Chinese, Japanese, Spanish, French, Italian, German, Portuguese).
• March, 2021: InfoXLM was accepted by NAACL 2021.
• December 29th, 2020: LayoutLMv2 is coming with the new SOTA on a wide varierty of document AI tasks, including DocVQA and SROIE leaderboard.
• October 8th, 2020: T-ULRv2 (aka InfoXLM) as the SOTA on the XTREME leaderboard. // Blog
• September, 2020: MiniLM was accepted by NeurIPS 2020.
• July 16, 2020: InfoXLM (Multilingual UniLM) arXiv
• June, 2020: UniLMv2 was accepted by ICML 2020; LayoutLM was accepted by KDD 2020.
• April 5, 2020: Multilingual MiniLM released!
• September, 2019: UniLMv1 was accepted by NeurIPS 2019.

Release

***** New July, 2021: BEiT release *****

• BEiT (June 15, 2021): BERT Pre-Training of Image Transformers. BEiT-large achieves state-of-the-art results on ADE20K (a big jump to 57.0 mIoU) for semantic segmentation. BEiT-large achieves state-of-the-art ImageNet top-1 accuracy (88.6%) under the setting without extra data other than ImageNet-22k. "BEiT: BERT Pre-Training of Image Transformers"

***** New June, 2021: LayoutXLM | AdaLM | MiniLMv2 release *****

• LayoutXLM (April 17, 2021): multimodal pre-training for multilingual visually-rich document understanding. The pre-trained LayoutXLM model has significantly outperformed the existing SOTA cross-lingual pre-trained models on the FUNSD and multilingual XFUND dataset including 7 languages (Chinese, Japanese, Spanish, French, Italian, German, Portuguese). "LayoutXLM: Multimodal Pre-training for Multilingual Visually-rich Document Understanding"
• AdaLM (June 2021): a simple yet effective approach for domain adaptation of pre-trained models. Biomedical specific pre-trained models are released. "Adapt-and-Distill: Developing Small, Fast and Effective Pretrained Language Models for Domains ACL 2021"
• MiniLMv2 (December, 2020): a simple yet effective task-agnostic knoweldge distillation method, namely multi-head self-attention relation distillation, for compressing large pre-trained Transformers into small and fast pre-trained models. MiniLMv2 significantly outperforms MiniLMv1. Both English and multilingual MiniLM models are released. "MiniLMv2: Multi-Head Self-Attention Relation Distillation for Compressing Pretrained Transformers ACL 2021"

***** New May, 2021: LayoutLMv2 | LayoutXLM release *****

• LayoutLM 2.0 (December 29, 2020): multimodal pre-training for visually-rich document understanding by leveraging text, layout and image information in a single framework. It is coming with new SOTA on a wide range of document understanding tasks, including FUNSD (0.7895 -> 0.8420), CORD (0.9493 -> 0.9601), SROIE (0.9524 -> 0.9781), Kleister-NDA (0.834 -> 0.852), RVL-CDIP (0.9443 -> 0.9564), and DocVQA (0.7295 -> 0.8672). "LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding ACL 2021"

***** February, 2020: UniLM v2 | MiniLM v1 | LayoutLM v1 | s2s-ft v1 release *****

• LayoutLM 1.0 (February 18, 2020): pre-trained models for document (image) understanding (e.g. receipts, forms, etc.) . It achieves new SOTA results in several downstream tasks, including form understanding (the FUNSD dataset from 70.72 to 79.27), receipt understanding (the ICDAR 2019 SROIE leaderboard from 94.02 to 95.24) and document image classification (the RVL-CDIP dataset from 93.07 to 94.42). "LayoutLM: Pre-training of Text and Layout for Document Image Understanding KDD 2020"
• s2s-ft 1.0 (February 26, 2020): A PyTorch package used to fine-tune pre-trained Transformers for sequence-to-sequence language generation. "s2s-ft: Fine-Tuning Pre-Trained Transformers for Sequence-to-Sequence Learning"
• MiniLM 1.0 (February 26, 2020): deep self-attention distillation is all you need (for task-agnostic knowledge distillation of pre-trained Transformers). MiniLM (12-layer, 384-hidden) achieves 2.7x speedup and comparable results over BERT-base (12-layer, 768-hidden) on NLU tasks as well as strong results on NLG tasks. The even smaller MiniLM (6-layer, 384-hidden) obtains 5.3x speedup and produces very competitive results. "MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers NeurIPS 2020"
• UniLM 2.0 (February 28, 2020): unified pre-training of bi-directional LM (via autoencoding) and sequence-to-sequence LM (via partially autoregressive) w/ Pseudo-Masked Language Model for language understanding and generation. UniLM v2 achieves new SOTA in a wide range of natural language understanding and generation tasks. "UniLMv2: Pseudo-Masked Language Models for Unified Language Model Pre-Training ICML 2020"

***** October 1st, 2019: UniLM v1 release *****

• UniLM v1 (September 30, 2019): the code and pre-trained models for the NeurIPS 2019 paper entitled "Unified Language Model Pre-training for Natural Language Understanding and Generation". UniLM (v1) achieves the new SOTA results in NLG (especially sequence-to-sequence generation) tasks, including abstractive summarization (the Gigaword and CNN/DM datasets), question generation (the SQuAD QG dataset), etc.

License

This project is licensed under the license found in the LICENSE file in the root directory of this source tree. Portions of the source code are based on the transformers project.

Microsoft Open Source Code of Conduct

Contact Information

For help or issues using UniLM AI models, please submit a GitHub issue.

For other communications related to UniLM AI, please contact Li Dong (lidong1@microsoft.com), Furu Wei (fuwei@microsoft.com).