{"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/bend-benchmarking-dna-language-models-on","title":"BEND: Benchmarking DNA Language Models on biologically meaningful tasks","arxiv_id":"2311.12570","date":"2023-11-21","proceeding":null,"authors":["Frederikke Isa Marin","Felix Teufel","Marc Horlacher","Dennis Madsen","Dennis Pultz","Ole Winther","Wouter Boomsma"],"abstract":"The genome sequence contains the blueprint for governing cellular processes. While the availability of genomes has vastly increased over the last decades, experimental annotation of the various functional, non-coding and regulatory elements encoded in the DNA sequence remains both expensive and challenging. This has sparked interest in unsupervised language modeling of genomic DNA, a paradigm that has seen great success for protein sequence data. Although various DNA language models have been proposed, evaluation tasks often differ between individual works, and might not fully recapitulate the fundamental challenges of genome annotation, including the length, scale and sparsity of the data. In this study, we introduce BEND, a Benchmark for DNA language models, featuring a collection of realistic and biologically meaningful downstream tasks defined on the human genome. We find that embeddings from current DNA LMs can approach performance of expert methods on some tasks, but only capture limited information about long-range features. BEND is available at https://github.com/frederikkemarin/BEND.","url_abs":"https://arxiv.org/abs/2311.12570v4","url_pdf":"https://arxiv.org/pdf/2311.12570v4.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":"bend-benchmarking-dna-language-models-on","repo_url":"https://github.com/frederikkemarin/bend","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"tasks":[{"task_slug":"benchmarking","task_name":"Benchmarking"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2311.12570","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2311.12570"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"deterministic:regex_extraction","url":"https://github.com/frederikkemarin/BEND","reach":{"status":"ok","spdx":"BSD-3-Clause"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/frederikkemarin/bend","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"summary":{"ran":8,"ran_fixture":1,"unverified":8},"by_repo_kind":{"official":{"samples":17,"ran":9,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"cdf5d52cced95cd6","entry":"collate_fn_pad_to_longest","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/utils/data_downstream.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/utils/data_downstream.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"cdf5d52cced95cd6"}},{"code_sha256_prefix":"be79703e8f09ed27","entry":"get_cls_by_name","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/models/gena_lm.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/models/gena_lm.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"be79703e8f09ed27"}},{"code_sha256_prefix":"0e86756b80a104e2","entry":"get_splits","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/io/sequtils.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/io/sequtils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"0e86756b80a104e2"}},{"code_sha256_prefix":"5dd28d20c60f0c19","entry":"mul_sum","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/models/hyena_dna.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/models/hyena_dna.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"5dd28d20c60f0c19"}},{"code_sha256_prefix":"69c309d20ae9b23c","entry":"multi_hot","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/io/sequtils.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/io/sequtils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"69c309d20ae9b23c"}},{"code_sha256_prefix":"38362b91aeed2794","entry":"pad_to_longest","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/utils/data_downstream.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/utils/data_downstream.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"38362b91aeed2794"}},{"code_sha256_prefix":"9e10ee856e01dce4","entry":"reverse_complement","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/io/sequtils.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/io/sequtils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"9e10ee856e01dce4"}},{"code_sha256_prefix":"2502f906145be7f6","entry":"rotate_half","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/models/gena_lm.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/models/gena_lm.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"2502f906145be7f6"}},{"code_sha256_prefix":"a41fa8c40b551ec1","entry":"unpad_input_only","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/models/dnabert2_padding.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/models/dnabert2_padding.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"a41fa8c40b551ec1"}},{"code_sha256_prefix":"9f62553a95737a2a","entry":"create_mixer_cls","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/models/hyena_dna.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/models/hyena_dna.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"9f62553a95737a2a"}},{"code_sha256_prefix":"a499a006a09e07f6","entry":"embed_dnabert","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/utils/embedders.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/utils/embedders.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"a499a006a09e07f6"}},{"code_sha256_prefix":"098e047c42924d30","entry":"embed_gpn","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/utils/embedders.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/utils/embedders.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"098e047c42924d30"}},{"code_sha256_prefix":"aa510fc4ec618e30","entry":"embed_nucleotide_transformer","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/utils/embedders.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/utils/embedders.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"aa510fc4ec618e30"}},{"code_sha256_prefix":"41b32f57fc8ec1e5","entry":"fftconv","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/models/hyena_dna.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/models/hyena_dna.py","link_basis":"plan_row","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"41b32f57fc8ec1e5"}},{"code_sha256_prefix":"4db905959447729d","entry":"load_tf_weights_in_bert","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/models/gena_lm.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/models/gena_lm.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"4db905959447729d"}},{"code_sha256_prefix":"b166dc11d0f54137","entry":"pad_input","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/models/dnabert2_padding.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/models/dnabert2_padding.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"b166dc11d0f54137"}},{"code_sha256_prefix":"ee341a843ab02d5f","entry":"unpad_input","repo":"frederikkemarin/BEND","repo_kind":"official","path":"bend/models/dnabert2_padding.py","file_url":"https://github.com/frederikkemarin/BEND/blob/HEAD/bend/models/dnabert2_padding.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"ee341a843ab02d5f"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}