{"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/axcell-automatic-extraction-of-results-from","title":"AxCell: Automatic Extraction of Results from Machine Learning Papers","arxiv_id":"2004.14356","date":"2020-04-29","proceeding":"EMNLP 2020 11","authors":["Marcin Kardas","Piotr Czapla","Pontus Stenetorp","Sebastian Ruder","Sebastian Riedel","Ross Taylor","Robert Stojnic"],"abstract":"Tracking progress in machine learning has become increasingly difficult with the recent explosion in the number of papers. In this paper, we present AxCell, an automatic machine learning pipeline for extracting results from papers. AxCell uses several novel components, including a table segmentation subtask, to learn relevant structural knowledge that aids extraction. When compared with existing methods, our approach significantly improves the state of the art for results extraction. We also release a structured, annotated dataset for training models for results extraction, and a dataset for evaluating the performance of models on this task. Lastly, we show the viability of our approach enables it to be used for semi-automated results extraction in production, suggesting our improvements make this task practically viable for the first time. Code is available on GitHub.","url_abs":"https://arxiv.org/abs/2004.14356v1","url_pdf":"https://arxiv.org/pdf/2004.14356v1.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":"axcell-automatic-extraction-of-results-from","repo_url":"https://github.com/paperswithcode/axcell","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"},{"task_slug":"scientific-results-extraction","task_name":"Scientific Results Extraction"}],"methods":[],"datasets_introduced":[{"slug":"arxivpapers","name":"ArxivPapers","full_name":""},{"slug":"linkedresults","name":"LinkedResults","full_name":""},{"slug":"pwc-leaderboards","name":"PWC Leaderboards","full_name":"Papers with Code Leaderboards"},{"slug":"segmentedtables","name":"SegmentedTables","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/scientific-results-extraction-on-nlp-tdms-exp","task":"Scientific Results Extraction","dataset":"NLP-TDMS (Exp, arXiv only)","model":"AxCell","rank_in_archive_order":1,"of":2,"metrics":{"Macro F1":"19.7","Macro Precision":"20.2","Macro Recall":"20.6","Micro F1":"25.8","Micro Precision":"27.4","Micro Recall":"24.4"},"uses_additional_data":false},{"leaderboard":"/sota/scientific-results-extraction-on-pwc","task":"Scientific Results Extraction","dataset":"PWC Leaderboards (restricted)","model":"AxCell","rank_in_archive_order":1,"of":1,"metrics":{"Macro F1":"21.1","Macro Precision":"24","Macro Recall":"21.8","Micro F1":"28.7","Micro Precision":"37.4","Micro Recall":"23.2"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2004.14356","atlas_url":"https://app.syntology.ai/?focus=2004.14356","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2004.14356"}},"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/paperswithcode/axcell","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":9,"unverified":5},"by_repo_kind":{"official":{"samples":14,"ran":9,"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":"53bc9344cc348810","entry":"annotations","repo":"paperswithcode/axcell","repo_kind":"official","path":"axcell/helpers/optimize.py","file_url":"https://github.com/paperswithcode/axcell/blob/HEAD/axcell/helpers/optimize.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"53bc9344cc348810"}},{"code_sha256_prefix":"83f38dae7103f51b","entry":"load_references","repo":"paperswithcode/axcell","repo_kind":"official","path":"axcell/helpers/cache.py","file_url":"https://github.com/paperswithcode/axcell/blob/HEAD/axcell/helpers/cache.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"83f38dae7103f51b"}},{"code_sha256_prefix":"ad584250d25d1fb1","entry":"load_structure","repo":"paperswithcode/axcell","repo_kind":"official","path":"axcell/helpers/cache.py","file_url":"https://github.com/paperswithcode/axcell/blob/HEAD/axcell/helpers/cache.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ad584250d25d1fb1"}},{"code_sha256_prefix":"e150c098756c6b8e","entry":"load_tags","repo":"paperswithcode/axcell","repo_kind":"official","path":"axcell/helpers/cache.py","file_url":"https://github.com/paperswithcode/axcell/blob/HEAD/axcell/helpers/cache.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"e150c098756c6b8e"}},{"code_sha256_prefix":"7fc6d3a39db0d9de","entry":"precision","repo":"paperswithcode/axcell","repo_kind":"official","path":"axcell/helpers/evaluate.py","file_url":"https://github.com/paperswithcode/axcell/blob/HEAD/axcell/helpers/evaluate.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"7fc6d3a39db0d9de"}},{"code_sha256_prefix":"4360ff7024678437","entry":"read_arxiv_papers","repo":"paperswithcode/axcell","repo_kind":"official","path":"axcell/helpers/datasets.py","file_url":"https://github.com/paperswithcode/axcell/blob/HEAD/axcell/helpers/datasets.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"4360ff7024678437"}},{"code_sha256_prefix":"f02f73aa554eecb3","entry":"recall","repo":"paperswithcode/axcell","repo_kind":"official","path":"axcell/helpers/evaluate.py","file_url":"https://github.com/paperswithcode/axcell/blob/HEAD/axcell/helpers/evaluate.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f02f73aa554eecb3"}},{"code_sha256_prefix":"f8b1fe756d77cd7e","entry":"ro_bind","repo":"paperswithcode/axcell","repo_kind":"official","path":"axcell/helpers/latex_converter.py","file_url":"https://github.com/paperswithcode/axcell/blob/HEAD/axcell/helpers/latex_converter.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f8b1fe756d77cd7e"}},{"code_sha256_prefix":"f05aaf2be87524df","entry":"rw_bind","repo":"paperswithcode/axcell","repo_kind":"official","path":"axcell/helpers/latex_converter.py","file_url":"https://github.com/paperswithcode/axcell/blob/HEAD/axcell/helpers/latex_converter.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f05aaf2be87524df"}},{"code_sha256_prefix":"86fb74ea9103b92b","entry":"estimate_context_noise","repo":"paperswithcode/axcell","repo_kind":"official","path":"axcell/helpers/optimize.py","file_url":"https://github.com/paperswithcode/axcell/blob/HEAD/axcell/helpers/optimize.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"86fb74ea9103b92b"}},{"code_sha256_prefix":"3387a9ba01a1d9a2","entry":"estimate_noises","repo":"paperswithcode/axcell","repo_kind":"official","path":"axcell/helpers/optimize.py","file_url":"https://github.com/paperswithcode/axcell/blob/HEAD/axcell/helpers/optimize.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"3387a9ba01a1d9a2"}},{"code_sha256_prefix":"b75165430b90d460","entry":"norm_score_str","repo":"paperswithcode/axcell","repo_kind":"official","path":"axcell/helpers/evaluate.py","file_url":"https://github.com/paperswithcode/axcell/blob/HEAD/axcell/helpers/evaluate.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b75165430b90d460"}},{"code_sha256_prefix":"02b6a005a3d5a4bb","entry":"read_tables_annotations","repo":"paperswithcode/axcell","repo_kind":"official","path":"axcell/helpers/datasets.py","file_url":"https://github.com/paperswithcode/axcell/blob/HEAD/axcell/helpers/datasets.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"02b6a005a3d5a4bb"}},{"code_sha256_prefix":"4767e70065da93cd","entry":"table_to_html","repo":"paperswithcode/axcell","repo_kind":"official","path":"axcell/helpers/jupyter.py","file_url":"https://github.com/paperswithcode/axcell/blob/HEAD/axcell/helpers/jupyter.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"4767e70065da93cd"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}