{"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/battrae-bidimensional-attention-based","title":"BattRAE: Bidimensional Attention-Based Recursive Autoencoders for Learning Bilingual Phrase Embeddings","arxiv_id":"1605.07874","date":"2016-05-25","proceeding":null,"authors":["Biao Zhang","Deyi Xiong","Jinsong Su"],"abstract":"In this paper, we propose a bidimensional attention based recursive\nautoencoder (BattRAE) to integrate clues and sourcetarget interactions at\nmultiple levels of granularity into bilingual phrase representations. We employ\nrecursive autoencoders to generate tree structures of phrases with embeddings\nat different levels of granularity (e.g., words, sub-phrases and phrases). Over\nthese embeddings on the source and target side, we introduce a bidimensional\nattention network to learn their interactions encoded in a bidimensional\nattention matrix, from which we extract two soft attention weight distributions\nsimultaneously. These weight distributions enable BattRAE to generate\ncompositive phrase representations via convolution. Based on the learned phrase\nrepresentations, we further use a bilinear neural model, trained via a\nmax-margin method, to measure bilingual semantic similarity. To evaluate the\neffectiveness of BattRAE, we incorporate this semantic similarity as an\nadditional feature into a state-of-the-art SMT system. Extensive experiments on\nNIST Chinese-English test sets show that our model achieves a substantial\nimprovement of up to 1.63 BLEU points on average over the baseline.","url_abs":"http://arxiv.org/abs/1605.07874v2","url_pdf":"http://arxiv.org/pdf/1605.07874v2.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":"battrae-bidimensional-attention-based","repo_url":"https://github.com/DeepLearnXMU/BattRAE","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"semantic-similarity","task_name":"Semantic Similarity"},{"task_slug":"semantic-textual-similarity","task_name":"Semantic Textual Similarity"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}