{"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/namer-non-autoregressive-modeling-for","title":"NAMER: Non-Autoregressive Modeling for Handwritten Mathematical Expression Recognition","arxiv_id":"2407.11380","date":"2024-07-16","proceeding":null,"authors":["Chenyu Liu","Jia Pan","Jinshui Hu","BaoCai Yin","Bing Yin","Mingjun Chen","Cong Liu","Jun Du","Qingfeng Liu"],"abstract":"Recently, Handwritten Mathematical Expression Recognition (HMER) has gained considerable attention in pattern recognition for its diverse applications in document understanding. Current methods typically approach HMER as an image-to-sequence generation task within an autoregressive (AR) encoder-decoder framework. However, these approaches suffer from several drawbacks: 1) a lack of overall language context, limiting information utilization beyond the current decoding step; 2) error accumulation during AR decoding; and 3) slow decoding speed. To tackle these problems, this paper makes the first attempt to build a novel bottom-up Non-AutoRegressive Modeling approach for HMER, called NAMER. NAMER comprises a Visual Aware Tokenizer (VAT) and a Parallel Graph Decoder (PGD). Initially, the VAT tokenizes visible symbols and local relations at a coarse level. Subsequently, the PGD refines all tokens and establishes connectivities in parallel, leveraging comprehensive visual and linguistic contexts. Experiments on CROHME 2014/2016/2019 and HME100K datasets demonstrate that NAMER not only outperforms the current state-of-the-art (SOTA) methods on ExpRate by 1.93%/2.35%/1.49%/0.62%, but also achieves significant speedups of 13.7x and 6.7x faster in decoding time and overall FPS, proving the effectiveness and efficiency of NAMER.","url_abs":"https://arxiv.org/abs/2407.11380v1","url_pdf":"https://arxiv.org/pdf/2407.11380v1.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":[],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"handwritten-mathmatical-expression","task_name":"Handwritten Mathmatical Expression Recognition"},{"task_slug":"document-understanding","task_name":"document understanding"}],"methods":[{"method_slug":"aware","method_name":"AWARE"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/handwritten-mathmatical-expression","task":"Handwritten Mathmatical Expression Recognition","dataset":"CROHME 2014","model":"NAMER","rank_in_archive_order":5,"of":17,"metrics":{"ExpRate":"60.51"},"uses_additional_data":false},{"leaderboard":"/sota/handwritten-mathmatical-expression-1","task":"Handwritten Mathmatical Expression Recognition","dataset":"CROHME 2016","model":"NAMER","rank_in_archive_order":4,"of":16,"metrics":{"ExpRate":"60.24"},"uses_additional_data":false},{"leaderboard":"/sota/handwritten-mathmatical-expression-2","task":"Handwritten Mathmatical Expression Recognition","dataset":"CROHME 2019","model":"NAMER","rank_in_archive_order":4,"of":14,"metrics":{"ExpRate":"61.72"},"uses_additional_data":false},{"leaderboard":"/sota/handwritten-mathmatical-expression-3","task":"Handwritten Mathmatical Expression Recognition","dataset":"HME100K","model":"NAMER","rank_in_archive_order":4,"of":12,"metrics":{"ExpRate":"68.52"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2407.11380","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}