{"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/alleviating-sequence-information-loss-with","title":"Alleviating Sequence Information Loss with Data Overlapping and Prime Batch Sizes","arxiv_id":"1909.08700","date":"2019-09-18","proceeding":"CONLL 2019 11","authors":["Noémien Kocher","Christian Scuito","Lorenzo Tarantino","Alexandros Lazaridis","Andreas Fischer","Claudiu Musat"],"abstract":"In sequence modeling tasks the token order matters, but this information can be partially lost due to the discretization of the sequence into data points. In this paper, we study the imbalance between the way certain token pairs are included in data points and others are not. We denote this a token order imbalance (TOI) and we link the partial sequence information loss to a diminished performance of the system as a whole, both in text and speech processing tasks. We then provide a mechanism to leverage the full token order information -Alleviated TOI- by iteratively overlapping the token composition of data points. For recurrent networks, we use prime numbers for the batch size to avoid redundancies when building batches from overlapped data points. The proposed method achieved state of the art performance in both text and speech related tasks.","url_abs":"https://arxiv.org/abs/1909.08700v1","url_pdf":"https://arxiv.org/pdf/1909.08700v1.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":"alleviating-sequence-information-loss-with","repo_url":"https://github.com/nkcr/overlap-ml","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"tasks":[{"task_slug":"language-modelling","task_name":"Language Modelling"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/language-modelling-on-wikitext-103","task":"Language Modelling","dataset":"WikiText-103","model":"AWD-LSTM-MoS + ATOI","rank_in_archive_order":74,"of":89,"metrics":{"Test perplexity":"32.85","Validation perplexity":"31.92"},"uses_additional_data":false},{"leaderboard":"/sota/language-modelling-on-wikitext-2","task":"Language Modelling","dataset":"WikiText-2","model":"AWD-LSTM + ATOI","rank_in_archive_order":30,"of":38,"metrics":{"Number of params":"33M","Test perplexity":"64.73","Validation perplexity":"67.47"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}