{"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/beyond-512-tokens-siamese-multi-depth","title":"Beyond 512 Tokens: Siamese Multi-depth Transformer-based Hierarchical Encoder for Long-Form Document Matching","arxiv_id":"2004.12297","date":"2020-04-26","proceeding":null,"authors":["Liu Yang","Mingyang Zhang","Cheng Li","Michael Bendersky","Marc Najork"],"abstract":"Many natural language processing and information retrieval problems can be formalized as the task of semantic matching. Existing work in this area has been largely focused on matching between short texts (e.g., question answering), or between a short and a long text (e.g., ad-hoc retrieval). Semantic matching between long-form documents, which has many important applications like news recommendation, related article recommendation and document clustering, is relatively less explored and needs more research effort. In recent years, self-attention based models like Transformers and BERT have achieved state-of-the-art performance in the task of text matching. These models, however, are still limited to short text like a few sentences or one paragraph due to the quadratic computational complexity of self-attention with respect to input text length. In this paper, we address the issue by proposing the Siamese Multi-depth Transformer-based Hierarchical (SMITH) Encoder for long-form document matching. Our model contains several innovations to adapt self-attention models for longer text input. In order to better capture sentence level semantic relations within a document, we pre-train the model with a novel masked sentence block language modeling task in addition to the masked word language modeling task used by BERT. Our experimental results on several benchmark datasets for long-form document matching show that our proposed SMITH model outperforms the previous state-of-the-art models including hierarchical attention, multi-depth attention-based hierarchical recurrent neural network, and BERT. Comparing to BERT based baselines, our model is able to increase maximum input text length from 512 to 2048. We will open source a Wikipedia based benchmark dataset, code and a pre-trained checkpoint to accelerate future research on long-form document matching.","url_abs":"https://arxiv.org/abs/2004.12297v2","url_pdf":"https://arxiv.org/pdf/2004.12297v2.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":"beyond-512-tokens-siamese-multi-depth","repo_url":"https://github.com/google-research/google-research/tree/master/smith","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"form","task_name":"Form"},{"task_slug":"information-retrieval","task_name":"Information Retrieval"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"natural-language-understanding","task_name":"Natural Language Understanding"},{"task_slug":"news-recommendation","task_name":"News Recommendation"},{"task_slug":"2048","task_name":"Playing the Game of 2048"},{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"text-matching","task_name":"Text Matching"}],"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":"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":"smith","method_name":"SMITH"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"weight-decay","method_name":"Weight Decay"},{"method_slug":"wordpiece","method_name":"WordPiece"}],"datasets_introduced":[],"methods_introduced":[{"slug":"smith","name":"SMITH","full_name":"Siamese Multi-depth Transformer-based Hierarchical Encoder"}],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2004.12297","atlas_url":"https://app.syntology.ai/?focus=2004.12297","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":"2004.12297"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2004.12297\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2004.12297"},"call":"get_concepts_for_paper(arxiv_id=\"2004.12297\")"}],"by_repo":[{"repo":"google-research/google-research","kind":"official","samples":0,"ran":0,"instrument":0,"unverified":0,"harvested_for_other_papers":24,"harvested_for":["1802.03268","2003.12022","2004.13912","2006.04678","2006.15055","2007.07314","2009.14794","2104.04657","2105.01237","2106.03314","2106.03760","2106.11905","2107.05768","2110.11346","2111.06888","2201.08896","2202.00512","2209.14881","2210.15458","2303.00732","2305.07011","2305.16381","2310.18893","2405.05349"],"harvested_for_calls":[{"arxiv_id":"1802.03268","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"1802.03268"},"call":"get_harvested_code_for_paper(arxiv_id=\"1802.03268\")"},{"arxiv_id":"2003.12022","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2003.12022"},"call":"get_harvested_code_for_paper(arxiv_id=\"2003.12022\")"},{"arxiv_id":"2004.13912","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2004.13912"},"call":"get_harvested_code_for_paper(arxiv_id=\"2004.13912\")"},{"arxiv_id":"2006.04678","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2006.04678"},"call":"get_harvested_code_for_paper(arxiv_id=\"2006.04678\")"},{"arxiv_id":"2006.15055","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2006.15055"},"call":"get_harvested_code_for_paper(arxiv_id=\"2006.15055\")"},{"arxiv_id":"2007.07314","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2007.07314"},"call":"get_harvested_code_for_paper(arxiv_id=\"2007.07314\")"},{"arxiv_id":"2009.14794","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2009.14794"},"call":"get_harvested_code_for_paper(arxiv_id=\"2009.14794\")"},{"arxiv_id":"2104.04657","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2104.04657"},"call":"get_harvested_code_for_paper(arxiv_id=\"2104.04657\")"},{"arxiv_id":"2105.01237","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2105.01237"},"call":"get_harvested_code_for_paper(arxiv_id=\"2105.01237\")"},{"arxiv_id":"2106.03314","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2106.03314"},"call":"get_harvested_code_for_paper(arxiv_id=\"2106.03314\")"},{"arxiv_id":"2106.03760","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2106.03760"},"call":"get_harvested_code_for_paper(arxiv_id=\"2106.03760\")"},{"arxiv_id":"2106.11905","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2106.11905"},"call":"get_harvested_code_for_paper(arxiv_id=\"2106.11905\")"},{"arxiv_id":"2107.05768","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2107.05768"},"call":"get_harvested_code_for_paper(arxiv_id=\"2107.05768\")"},{"arxiv_id":"2110.11346","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2110.11346"},"call":"get_harvested_code_for_paper(arxiv_id=\"2110.11346\")"},{"arxiv_id":"2111.06888","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2111.06888"},"call":"get_harvested_code_for_paper(arxiv_id=\"2111.06888\")"},{"arxiv_id":"2201.08896","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2201.08896"},"call":"get_harvested_code_for_paper(arxiv_id=\"2201.08896\")"},{"arxiv_id":"2202.00512","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2202.00512"},"call":"get_harvested_code_for_paper(arxiv_id=\"2202.00512\")"},{"arxiv_id":"2209.14881","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2209.14881"},"call":"get_harvested_code_for_paper(arxiv_id=\"2209.14881\")"},{"arxiv_id":"2210.15458","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2210.15458"},"call":"get_harvested_code_for_paper(arxiv_id=\"2210.15458\")"},{"arxiv_id":"2303.00732","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2303.00732"},"call":"get_harvested_code_for_paper(arxiv_id=\"2303.00732\")"},{"arxiv_id":"2305.07011","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2305.07011"},"call":"get_harvested_code_for_paper(arxiv_id=\"2305.07011\")"},{"arxiv_id":"2305.16381","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2305.16381"},"call":"get_harvested_code_for_paper(arxiv_id=\"2305.16381\")"},{"arxiv_id":"2310.18893","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2310.18893"},"call":"get_harvested_code_for_paper(arxiv_id=\"2310.18893\")"},{"arxiv_id":"2405.05349","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2405.05349"},"call":"get_harvested_code_for_paper(arxiv_id=\"2405.05349\")"}]}],"by_repo_harvested_for_is":"harvested_for: the arXiv ids (as Syntology's graph holds them) of every paper whose samples Syntology's graph links to this repository, in arXiv-id order, complete; the page names the first 3 and says how many more. A link only: it does not say the papers are the same work, that one copied the other, or that the code implements this paper's method; harvested_for_calls gives the get_harvested_code_for_paper call for each arXiv-shaped id among them","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":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"d57747b93813867b71e58d9cbc80276391c95600022e541c285a97966a916140","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}