{"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/transformer-xh-multi-evidence-reasoning-with","title":"Transformer-XH: Multi-Evidence Reasoning with eXtra Hop Attention","arxiv_id":null,"date":"2020-05-01","proceeding":"ICLR 2020 1","authors":["Chen Zhao","Chenyan Xiong","Corby Rosset","Xia Song","Paul Bennett","Saurabh Tiwary"],"abstract":"Transformers have achieved new heights modeling natural language as a sequence of text tokens. However, in many real world scenarios, textual data inherently exhibits structures beyond a linear sequence such as trees and graphs; many tasks require reasoning with evidence scattered across multiple pieces of texts. This paper presents Transformer-XH, which uses eXtra Hop attention to enable intrinsic modeling of structured texts in a fully data-driven way. Its new attention mechanism naturally “hops” across the connected text sequences in addition to attending over tokens within each sequence. Thus, Transformer-XH better conducts joint multi-evidence reasoning by propagating information between documents and constructing global contextualized representations. On multi-hop question answering, Transformer-XH leads to a simpler multi-hop QA system which outperforms previous state-of-the-art on the HotpotQA FullWiki setting. On FEVER fact verification, applying Transformer-XH provides state-of-the-art accuracy and excels on claims whose verification requires multiple evidence.","url_abs":"https://openreview.net/forum?id=r1eIiCNYwS","url_pdf":"https://openreview.net/pdf?id=r1eIiCNYwS","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":"transformer-xh-multi-evidence-reasoning-with","repo_url":"https://github.com/microsoft/Transformer-XH","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"fact-verification","task_name":"Fact Verification"},{"task_slug":"multi-hop-question-answering","task_name":"Multi-hop Question Answering"},{"task_slug":"question-answering","task_name":"Question Answering"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/question-answering-on-hotpotqa","task":"Question Answering","dataset":"HotpotQA","model":"Transformer-XH-final","rank_in_archive_order":42,"of":72,"metrics":{"ANS-EM":"0.516","ANS-F1":"0.641","JOINT-EM":"0.261","JOINT-F1":"0.513","SUP-EM":"0.409","SUP-F1":"0.714"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}