{"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/resolving-the-scope-of-speculation-and","title":"Resolving the Scope of Speculation and Negation using Transformer-Based Architectures","arxiv_id":"2001.02885","date":"2020-01-09","proceeding":null,"authors":["Benita Kathleen Britto","Aditya Khandelwal"],"abstract":"Speculation is a naturally occurring phenomena in textual data, forming an integral component of many systems, especially in the biomedical information retrieval domain. Previous work addressing cue detection and scope resolution (the two subtasks of speculation detection) have ranged from rule-based systems to deep learning-based approaches. In this paper, we apply three popular transformer-based architectures, BERT, XLNet and RoBERTa to this task, on two publicly available datasets, BioScope Corpus and SFU Review Corpus, reporting substantial improvements over previously reported results (by at least 0.29 F1 points on cue detection and 4.27 F1 points on scope resolution). We also experiment with joint training of the model on multiple datasets, which outperforms the single dataset training approach by a good margin. We observe that XLNet consistently outperforms BERT and RoBERTa, contrary to results on other benchmark datasets. To confirm this observation, we apply XLNet and RoBERTa to negation detection and scope resolution, reporting state-of-the-art results on negation scope resolution for the BioScope Corpus (increase of 3.16 F1 points on the BioScope Full Papers, 0.06 F1 points on the BioScope Abstracts) and the SFU Review Corpus (increase of 0.3 F1 points).","url_abs":"https://arxiv.org/abs/2001.02885v1","url_pdf":"https://arxiv.org/pdf/2001.02885v1.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":"resolving-the-scope-of-speculation-and","repo_url":"https://github.com/adityak6798/Transformers-For-Negation-and-Speculation","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"biomedical-information-retrieval","task_name":"Biomedical Information Retrieval"},{"task_slug":"information-retrieval","task_name":"Information Retrieval"},{"task_slug":"negation","task_name":"Negation"},{"task_slug":"negation-detection","task_name":"Negation Detection"},{"task_slug":"negation-scope-resolution","task_name":"Negation Scope Resolution"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"speculation-detection","task_name":"Speculation Detection"},{"task_slug":"speculation-scope-resolution","task_name":"Speculation Scope Resolution"}],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bert","method_name":"BERT"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-linear-decay","method_name":"Linear Warmup With Linear Decay"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"roberta","method_name":"RoBERTa"},{"method_slug":"sentencepiece","method_name":"SentencePiece"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"weight-decay","method_name":"Weight Decay"},{"method_slug":"wordpiece","method_name":"WordPiece"},{"method_slug":"xlnet","method_name":"XLNet"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/negation-scope-resolution-on-sem-2012-shared","task":"Negation Scope Resolution","dataset":"*sem 2012 Shared Task: Sherlock Dataset","model":"RoBERTa","rank_in_archive_order":3,"of":3,"metrics":{"F1":"91.59"},"uses_additional_data":true},{"leaderboard":"/sota/negation-scope-resolution-on-bioscope","task":"Negation Scope Resolution","dataset":"BioScope : Abstracts","model":"XLNet","rank_in_archive_order":2,"of":3,"metrics":{"F1":"95.74"},"uses_additional_data":true},{"leaderboard":"/sota/negation-scope-resolution-on-bioscope-full","task":"Negation Scope Resolution","dataset":"BioScope : Full Papers","model":"XLNet","rank_in_archive_order":1,"of":2,"metrics":{"F1":"94.4"},"uses_additional_data":true},{"leaderboard":"/sota/negation-scope-resolution-on-sfu-review","task":"Negation Scope Resolution","dataset":"SFU Review Corpus","model":"XLNet","rank_in_archive_order":1,"of":2,"metrics":{"F1":"91.25"},"uses_additional_data":true},{"leaderboard":"/sota/speculation-scope-resolution-on-bioscope","task":"Speculation Scope Resolution","dataset":"BioScope : Abstracts","model":"XLNet","rank_in_archive_order":2,"of":2,"metrics":{"F1":"97.87"},"uses_additional_data":true},{"leaderboard":"/sota/speculation-scope-resolution-on-bioscope-full","task":"Speculation Scope Resolution","dataset":"BioScope : Full Papers","model":"XLNet","rank_in_archive_order":1,"of":1,"metrics":{"F1":"96.91"},"uses_additional_data":true},{"leaderboard":"/sota/speculation-scope-resolution-on-sfu-review","task":"Speculation Scope Resolution","dataset":"SFU Review Corpus","model":"XLNet","rank_in_archive_order":1,"of":1,"metrics":{"F1":"91.00"},"uses_additional_data":true}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}