{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/should-we-be-pre-training-an-argument-for-end","title":"Should We Be Pre-training? An Argument for End-task Aware Training as an Alternative","arxiv_id":"2109.07437","date":"2021-09-15","proceeding":"ICLR 2022 4","authors":["Lucio M. Dery","Paul Michel","Ameet Talwalkar","Graham Neubig"],"abstract":"In most settings of practical concern, machine learning practitioners know in advance what end-task they wish to boost with auxiliary tasks. However, widely used methods for leveraging auxiliary data like pre-training and its continued-pretraining variant are end-task agnostic: they rarely, if ever, exploit knowledge of the target task. We study replacing end-task agnostic continued training of pre-trained language models with end-task aware training of said models. We argue that for sufficiently important end-tasks, the benefits of leveraging auxiliary data in a task-aware fashion can justify forgoing the traditional approach of obtaining generic, end-task agnostic representations as with (continued) pre-training. On three different low-resource NLP tasks from two domains, we demonstrate that multi-tasking the end-task and auxiliary objectives results in significantly better downstream task performance than the widely-used task-agnostic continued pre-training paradigm of Gururangan et al. (2020). We next introduce an online meta-learning algorithm that learns a set of multi-task weights to better balance among our multiple auxiliary objectives, achieving further improvements on end-task performance and data efficiency.","url_abs":"https://arxiv.org/abs/2109.07437v2","url_pdf":"https://arxiv.org/pdf/2109.07437v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"should-we-be-pre-training-an-argument-for-end","repo_url":"https://github.com/ldery/tartan","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"should-we-be-pre-training-an-argument-for-end","repo_url":"https://github.com/ldery/automating-auxiliary-learning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"meta-learning","task_name":"Meta-Learning"}],"methods":[{"method_slug":"aware","method_name":"AWARE"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2109.07437","atlas_url":"https://app.syntology.ai/?focus=2109.07437","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2109.07437"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ldery/automating-auxiliary-learning","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ldery/tartan","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":7},"by_repo_kind":{"official":{"samples":7,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"9d6a95b950631903","entry":"calc_norm","repo":"ldery/tartan","repo_kind":"official","path":"simple_vision_tartan/tartan_utils.py","file_url":"https://github.com/ldery/tartan/blob/HEAD/simple_vision_tartan/tartan_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9d6a95b950631903"}},{"code_sha256_prefix":"c54b69f5721172e3","entry":"cosine","repo":"ldery/tartan","repo_kind":"official","path":"simple_vision_tartan/tartan_utils.py","file_url":"https://github.com/ldery/tartan/blob/HEAD/simple_vision_tartan/tartan_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c54b69f5721172e3"}},{"code_sha256_prefix":"6502cf86619c912c","entry":"dot_prod","repo":"ldery/tartan","repo_kind":"official","path":"simple_vision_tartan/tartan_utils.py","file_url":"https://github.com/ldery/tartan/blob/HEAD/simple_vision_tartan/tartan_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6502cf86619c912c"}},{"code_sha256_prefix":"5619edddef62bb1e","entry":"get_model","repo":"ldery/tartan","repo_kind":"official","path":"simple_vision_tartan/run_tartan.py","file_url":"https://github.com/ldery/tartan/blob/HEAD/simple_vision_tartan/run_tartan.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5619edddef62bb1e"}},{"code_sha256_prefix":"0b1b624b8ddb1e3b","entry":"get_optimizer","repo":"ldery/tartan","repo_kind":"official","path":"simple_vision_tartan/run_tartan.py","file_url":"https://github.com/ldery/tartan/blob/HEAD/simple_vision_tartan/run_tartan.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0b1b624b8ddb1e3b"}},{"code_sha256_prefix":"20afabbb41adfa6e","entry":"set_random_seed","repo":"ldery/tartan","repo_kind":"official","path":"simple_vision_tartan/run_tartan.py","file_url":"https://github.com/ldery/tartan/blob/HEAD/simple_vision_tartan/run_tartan.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"20afabbb41adfa6e"}},{"code_sha256_prefix":"2ddbc2d5391caae1","entry":"to_tensor","repo":"ldery/tartan","repo_kind":"official","path":"simple_vision_tartan/data.py","file_url":"https://github.com/ldery/tartan/blob/HEAD/simple_vision_tartan/data.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2ddbc2d5391caae1"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}