{"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/end-to-end-information-extraction-without","title":"End-to-End Information Extraction without Token-Level Supervision","arxiv_id":"1707.04913","date":"2017-07-16","proceeding":"WS 2017 9","authors":["Rasmus Berg Palm","Dirk Hovy","Florian Laws","Ole Winther"],"abstract":"Most state-of-the-art information extraction approaches rely on token-level\nlabels to find the areas of interest in text. Unfortunately, these labels are\ntime-consuming and costly to create, and consequently, not available for many\nreal-life IE tasks. To make matters worse, token-level labels are usually not\nthe desired output, but just an intermediary step. End-to-end (E2E) models,\nwhich take raw text as input and produce the desired output directly, need not\ndepend on token-level labels. We propose an E2E model based on pointer\nnetworks, which can be trained directly on pairs of raw input and output text.\nWe evaluate our model on the ATIS data set, MIT restaurant corpus and the MIT\nmovie corpus and compare to neural baselines that do use token-level labels. We\nachieve competitive results, within a few percentage points of the baselines,\nshowing the feasibility of E2E information extraction without the need for\ntoken-level labels. This opens up new possibilities, as for many tasks\ncurrently addressed by human extractors, raw input and output data are\navailable, but not token-level labels.","url_abs":"http://arxiv.org/abs/1707.04913v1","url_pdf":"http://arxiv.org/pdf/1707.04913v1.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":"end-to-end-information-extraction-without","repo_url":"https://github.com/rasmusbergpalm/e2e-ie-release","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}