{"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/eliminet-a-model-for-eliminating-options-for-1","title":"ElimiNet: A Model for Eliminating Options for Reading Comprehension with Multiple Choice Questions","arxiv_id":"1904.02651","date":"2019-04-04","proceeding":"ICLR 2018 1","authors":["Soham Parikh","Ananya B. Sai","Preksha Nema","Mitesh M. Khapra"],"abstract":"The task of Reading Comprehension with Multiple Choice Questions, requires a\nhuman (or machine) to read a given passage, question pair and select one of the\nn given options. The current state of the art model for this task first\ncomputes a question-aware representation for the passage and then selects the\noption which has the maximum similarity with this representation. However, when\nhumans perform this task they do not just focus on option selection but use a\ncombination of elimination and selection. Specifically, a human would first try\nto eliminate the most irrelevant option and then read the passage again in the\nlight of this new information (and perhaps ignore portions corresponding to the\neliminated option). This process could be repeated multiple times till the\nreader is finally ready to select the correct option. We propose ElimiNet, a\nneural network-based model which tries to mimic this process. Specifically, it\nhas gates which decide whether an option can be eliminated given the passage,\nquestion pair and if so it tries to make the passage representation orthogonal\nto this eliminated option (akin to ignoring portions of the passage\ncorresponding to the eliminated option). The model makes multiple rounds of\npartial elimination to refine the passage representation and finally uses a\nselection module to pick the best option. We evaluate our model on the recently\nreleased large scale RACE dataset and show that it outperforms the current\nstate of the art model on 7 out of the $13$ question types in this dataset.\nFurther, we show that taking an ensemble of our elimination-selection based\nmethod with a selection based method gives us an improvement of 3.1% over the\nbest-reported performance on this dataset.","url_abs":"http://arxiv.org/abs/1904.02651v1","url_pdf":"http://arxiv.org/pdf/1904.02651v1.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":"eliminet-a-model-for-eliminating-options-for-1","repo_url":"https://github.com/sohamparikh94/ElimiNet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"multiple-choice","task_name":"Multiple-choice"},{"task_slug":"reading-comprehension","task_name":"Reading Comprehension"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}