{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/hierarchical-learning-for-generation-with","title":"Hierarchical Learning for Generation with Long Source Sequences","arxiv_id":"2104.07545","date":"2021-04-15","proceeding":null,"authors":["Tobias Rohde","Xiaoxia Wu","Yinhan Liu"],"abstract":"One of the challenges for current sequence to sequence (seq2seq) models is processing long sequences, such as those in summarization and document level machine translation tasks. These tasks require the model to reason at the token level as well as the sentence and paragraph level. We design and study a new Hierarchical Attention Transformer-based architecture (HAT) that outperforms standard Transformers on several sequence to sequence tasks. Furthermore, our model achieves state-of-the-art ROUGE scores on four summarization tasks, including PubMed, arXiv, CNN/DM, SAMSum, and AMI. Our model outperforms document-level machine translation baseline on the WMT20 English to German translation task. We investigate what the hierarchical layers learn by visualizing the hierarchical encoder-decoder attention. Finally, we study hierarchical learning on encoder-only pre-training and analyze its performance on classification tasks.","url_abs":"https://arxiv.org/abs/2104.07545v2","url_pdf":"https://arxiv.org/pdf/2104.07545v2.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":"document-level-machine-translation","task_name":"Document Level Machine Translation"},{"task_slug":"document-summarization","task_name":"Document Summarization"},{"task_slug":"document-translation","task_name":"Document Translation"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"reading-comprehension","task_name":"Reading Comprehension"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"text-summarization","task_name":"Text Summarization"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/document-summarization-on-cnn-daily-mail","task":"Document Summarization","dataset":"CNN / Daily Mail","model":"HAT-BART","rank_in_archive_order":5,"of":26,"metrics":{"ROUGE-1":"44.48","ROUGE-2":"21.31","ROUGE-L":"41.52"},"uses_additional_data":false},{"leaderboard":"/sota/reading-comprehension-on-race","task":"Reading Comprehension","dataset":"RACE","model":"HAT (Encoder)","rank_in_archive_order":10,"of":24,"metrics":{"Accuracy":"67.3"},"uses_additional_data":false},{"leaderboard":"/sota/text-summarization-on-ami","task":"Text Summarization","dataset":"AMI","model":"HAT-CNNDM","rank_in_archive_order":1,"of":1,"metrics":{"ROUGE-1":"52.27","ROUGE-2":"20.15","ROUGE-L":"50.57"},"uses_additional_data":false},{"leaderboard":"/sota/text-summarization-on-arxiv","task":"Text Summarization","dataset":"Arxiv HEP-TH citation graph","model":"HAT-BART","rank_in_archive_order":13,"of":28,"metrics":{"ROUGE-1":"46.74","ROUGE-2":"19.19","ROUGE-L":"42.2"},"uses_additional_data":true},{"leaderboard":"/sota/text-summarization-on-pubmed-1","task":"Text Summarization","dataset":"Pubmed","model":"HAT-BART","rank_in_archive_order":9,"of":29,"metrics":{"ROUGE-1":"48.25","ROUGE-2":"21.35","ROUGE-L":"36.69"},"uses_additional_data":true},{"leaderboard":"/sota/text-summarization-on-samsum-corpus","task":"Text Summarization","dataset":"SAMSum","model":"HAT-CNNDM","rank_in_archive_order":7,"of":12,"metrics":{"ROUGE-1":"53.01","ROUGE-2":"28.27"},"uses_additional_data":false},{"leaderboard":"/sota/text-summarization-on-samsum-corpus","task":"Text Summarization","dataset":"SAMSum","model":"HAT-CNNDM RL","rank_in_archive_order":11,"of":12,"metrics":{"ROUGE-L":"48.84"},"uses_additional_data":false},{"leaderboard":"/sota/text-summarization-on-x-sum","task":"Text Summarization","dataset":"X-Sum","model":"HAT-BART","rank_in_archive_order":6,"of":18,"metrics":{"ROUGE-1":"45.92","ROUGE-2":"22.79"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2104.07545","atlas_url":"https://app.syntology.ai/?focus=2104.07545","mcp":null,"developers":"https://syntology.ai/developers","held":true,"reason":"arxiv_id_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"agent_calls":[{"tool":"get_citation_path","arguments":{"paper_1":"2104.07545"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2104.07545\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2104.07545"},"call":"get_concepts_for_paper(arxiv_id=\"2104.07545\")"}],"by_repo":[],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":"the archive holds no code link for this paper","arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"090a7df2d88110891ed95613590dab7850c7a51ef229743d23aa8a03541c3966","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}