While deep and large pre-trained models are the state-of-the-art for various natural language processing tasks, their huge size poses significant challenges for practical uses in resource constrained settings. Recent works in knowledge distillation propose task-agnostic as well as task-specific methods to compress these models, with task-specific ones often yielding higher compression rate. In this work, we develop a new task-agnostic distillation framework XtremeDistilTransformers that leverages the advantage of task-specific methods for learning a small universal model that can be applied to arbitrary tasks and languages. To this end, we study the transferability of several source tasks, augmentation resources and model architecture for distillation. We evaluate our model performance on multiple tasks, including the General Language Understanding Evaluation (GLUE) benchmark, SQuAD question answering dataset and a massive multi-lingual NER dataset with 41 languages. We release three distilled task-agnostic checkpoints with 13MM, 22MM and 33MM parameters obtaining SOTA performance in several tasks.<\/p>\n","protected":false},"excerpt":{"rendered":"
While deep and large pre-trained models are the state-of-the-art for various natural language processing tasks, their huge size poses significant challenges for practical uses in resource constrained settings. Recent works in knowledge distillation propose task-agnostic as well as task-specific methods to compress these models, with task-specific ones often yielding higher compression rate. In this work, […]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","footnotes":""},"msr-content-type":[3],"msr-research-highlight":[],"research-area":[13545],"msr-publication-type":[193715],"msr-product-type":[],"msr-focus-area":[],"msr-platform":[],"msr-download-source":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-754180","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-human-language-technologies","msr-locale-en_us"],"msr_publishername":"","msr_edition":"","msr_affiliation":"","msr_published_date":"2021-6","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"ArXiv","msr_volume":"","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"","msr_how_published":"","msr_notes":"arXiv:2106.04563","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":1,"msr_main_download":"","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"url","viewUrl":"false","id":"false","title":"https:\/\/arxiv.org\/pdf\/2106.04563.pdf","label_id":"243109","label":0}],"msr_related_uploader":"","msr_attachments":[],"msr-author-ordering":[{"type":"user_nicename","value":"Subhabrata (Subho) Mukherjee","user_id":38308,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Subhabrata (Subho) Mukherjee"},{"type":"user_nicename","value":"Ahmed H. Awadallah","user_id":31979,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Ahmed H. Awadallah"},{"type":"user_nicename","value":"Jianfeng Gao","user_id":32246,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Jianfeng Gao"}],"msr_impact_theme":[],"msr_research_lab":[199565],"msr_event":[],"msr_group":[392600,644373,702211],"msr_project":[804847,675633],"publication":[],"video":[],"download":[838564],"msr_publication_type":"article","related_content":{"projects":[{"ID":804847,"post_title":"Reducing AI's Carbon Footprint","post_name":"reducing-ais-carbon-footprint","post_type":"msr-project","post_date":"2022-05-24 08:56:55","post_modified":"2024-01-16 11:11:59","post_status":"publish","permalink":"https:\/\/www.microsoft.com\/en-us\/research\/project\/reducing-ais-carbon-footprint\/","post_excerpt":"This project develops techniques that enable AI to use computing infrastructure more efficiently. The goals are to maintain predictive accuracy while reducing carbon emissions, whether embodied in manufactured hardware, or produced from electricity usage when green energy is not available.","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/804847"}]}},{"ID":675633,"post_title":"Knowledge Distillation","post_name":"xtreme-knowledge-distillation","post_type":"msr-project","post_date":"2020-07-15 19:55:49","post_modified":"2021-06-23 18:09:19","post_status":"publish","permalink":"https:\/\/www.microsoft.com\/en-us\/research\/project\/xtreme-knowledge-distillation\/","post_excerpt":"Modern machine learning applications have enjoyed a great boost utilizing deep and large neural network models, allowing them to achieve state-of-the-art results on a wide range of tasks such as question-answering, conversational AI, search and recommendation. A significant challenge facing practitioners is how to deploy these huge models in practice. Recent pre-trained language models like Turing-NLG and GPT-3 boast of a massive 17 billion and 175 billion parameters, respectively. Although they obtain superior performance in…","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/675633"}]}}]},"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/754180"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":2,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/754180\/revisions"}],"predecessor-version":[{"id":754189,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/754180\/revisions\/754189"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=754180"}],"wp:term":[{"taxonomy":"msr-content-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-content-type?post=754180"},{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=754180"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=754180"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=754180"},{"taxonomy":"msr-product-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-product-type?post=754180"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=754180"},{"taxonomy":"msr-platform","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-platform?post=754180"},{"taxonomy":"msr-download-source","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-download-source?post=754180"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=754180"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=754180"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=754180"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=754180"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=754180"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=754180"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=754180"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}